Loading [MathJax]/jax/output/SVG/jax.js
Review

Artificial intelligence generated content (AIGC) in medicine: A narrative review


  • Received: 27 September 2023 Revised: 05 December 2023 Accepted: 13 December 2023 Published: 02 January 2024
  • Recently, artificial intelligence generated content (AIGC) has been receiving increased attention and is growing exponentially. AIGC is generated based on the intentional information extracted from human-provided instructions by generative artificial intelligence (AI) models. AIGC quickly and automatically generates large amounts of high-quality content. Currently, there is a shortage of medical resources and complex medical procedures in medicine. Due to its characteristics, AIGC can help alleviate these problems. As a result, the application of AIGC in medicine has gained increased attention in recent years. Therefore, this paper provides a comprehensive review on the recent state of studies involving AIGC in medicine. First, we present an overview of AIGC. Furthermore, based on recent studies, the application of AIGC in medicine is reviewed from two aspects: medical image processing and medical text generation. The basic generative AI models, tasks, target organs, datasets and contribution of studies are considered and summarized. Finally, we also discuss the limitations and challenges faced by AIGC and propose possible solutions with relevant studies. We hope this review can help readers understand the potential of AIGC in medicine and obtain some innovative ideas in this field.

    Citation: Liangjing Shao, Benshuang Chen, Ziqun Zhang, Zhen Zhang, Xinrong Chen. Artificial intelligence generated content (AIGC) in medicine: A narrative review[J]. Mathematical Biosciences and Engineering, 2024, 21(1): 1672-1711. doi: 10.3934/mbe.2024073

    Related Papers:

    [1] Carlos Lizama, Marina Murillo-Arcila, Macarena Trujillo . Fractional Beer-Lambert law in laser heating of biological tissue. AIMS Mathematics, 2022, 7(8): 14444-14459. doi: 10.3934/math.2022796
    [2] Kiran Sajjan, Nehad Ali Shah, N. Ameer Ahammad, C.S.K. Raju, M. Dinesh Kumar, Wajaree Weera . Nonlinear Boussinesq and Rosseland approximations on 3D flow in an interruption of Ternary nanoparticles with various shapes of densities and conductivity properties. AIMS Mathematics, 2022, 7(10): 18416-18449. doi: 10.3934/math.20221014
    [3] Ibrahim-Elkhalil Ahmed, Ahmed E. Abouelregal, Doaa Atta, Meshari Alesemi . A fractional dual-phase-lag thermoelastic model for a solid half-space with changing thermophysical properties involving two-temperature and non-singular kernels. AIMS Mathematics, 2024, 9(3): 6964-6992. doi: 10.3934/math.2024340
    [4] Ahmed E. Abouelregal, Khalil M. Khalil, Wael W. Mohammed, Doaa Atta . Thermal vibration in rotating nanobeams with temperature-dependent due to exposure to laser irradiation. AIMS Mathematics, 2022, 7(4): 6128-6152. doi: 10.3934/math.2022341
    [5] Cheng-Hung Huang, Tsung-Yi Lee . Predicting the optimal heating function for uniform exit temperature in a pipe flow. AIMS Mathematics, 2024, 9(1): 1997-2021. doi: 10.3934/math.2024099
    [6] Abdelkader Moumen, Fares Yazid, Fatima Siham Djeradi, Moheddine Imsatfia, Tayeb Mahrouz, Keltoum Bouhali . The influence of damping on the asymptotic behavior of solution for laminated beam. AIMS Mathematics, 2024, 9(8): 22602-22626. doi: 10.3934/math.20241101
    [7] Dumitru Baleanu, Kamyar Hosseini, Soheil Salahshour, Khadijeh Sadri, Mohammad Mirzazadeh, Choonkil Park, Ali Ahmadian . The (2+1)-dimensional hyperbolic nonlinear Schrödinger equation and its optical solitons. AIMS Mathematics, 2021, 6(9): 9568-9581. doi: 10.3934/math.2021556
    [8] Nadeem Abbas, Wasfi Shatanawi, Taqi A. M. Shatnawi . Innovation of prescribe conditions for radiative Casson micropolar hybrid nanofluid flow with inclined MHD over a stretching sheet/cylinder. AIMS Mathematics, 2025, 10(2): 3561-3580. doi: 10.3934/math.2025164
    [9] Bauyrzhan Derbissaly, Makhmud Sadybekov . Inverse source problem for multi-term time-fractional diffusion equation with nonlocal boundary conditions. AIMS Mathematics, 2024, 9(4): 9969-9988. doi: 10.3934/math.2024488
    [10] Mohammed Alrehili . Managing heat transfer effectiveness in a Darcy medium with a vertically non-linear stretching surface through the flow of an electrically conductive non-Newtonian nanofluid. AIMS Mathematics, 2024, 9(4): 9195-9210. doi: 10.3934/math.2024448
  • Recently, artificial intelligence generated content (AIGC) has been receiving increased attention and is growing exponentially. AIGC is generated based on the intentional information extracted from human-provided instructions by generative artificial intelligence (AI) models. AIGC quickly and automatically generates large amounts of high-quality content. Currently, there is a shortage of medical resources and complex medical procedures in medicine. Due to its characteristics, AIGC can help alleviate these problems. As a result, the application of AIGC in medicine has gained increased attention in recent years. Therefore, this paper provides a comprehensive review on the recent state of studies involving AIGC in medicine. First, we present an overview of AIGC. Furthermore, based on recent studies, the application of AIGC in medicine is reviewed from two aspects: medical image processing and medical text generation. The basic generative AI models, tasks, target organs, datasets and contribution of studies are considered and summarized. Finally, we also discuss the limitations and challenges faced by AIGC and propose possible solutions with relevant studies. We hope this review can help readers understand the potential of AIGC in medicine and obtain some innovative ideas in this field.



    Contact and non-contact temperature measurement methods are commonly used to measure internal temperatures in aerospace and aviation engines. Contact methods, which are suitable below 1600℃, include thermocouples, temperature-indicating paints, crystal-based techniques, and optical fibers [1]. Non-contact methods include thermal imaging cameras and multispectral approaches [2,3]. Though effective, contact methods are limited to shorter-term applications in oxidizing environments at temperatures exceeding 1600 ℃. Measuring rakes made of platinum-rhodium thermocouples, for example, can be used only for brief durations in oxidizing environments above 1600 ℃, such as temperature measurements in engine combustion chambers [4,5,6]; furthermore, the precious metals necessary to fabricate these thermocouples are very costly. Temperature-indicating paints and crystal temperature measurement techniques are effective but only within engine combustion chambers and blades at maximum environmental temperatures and they cannot be used to measure temperature changes in real-time [7,8,9]. Optical fibers are significantly affected by ambient light in the testing environment, rendering measurements inaccurate when temperatures exceed 1300 ℃ [10]. Non-contact measurement methods also have significant limitations. Changes in emissivity, as well as factors, such as moisture and dust in the exhaust gas, can affect their accuracy [11].

    Traditional measurement methods do not perform effectively in high-temperature oxidizing environments, such as those of aircraft engines, where extreme temperatures can cause turbine blades and other components to burn out. Accurately measuring internal temperatures in aerospace engines is crucial for optimizing structural designs, minimizing redundancy, and improving overall performance. There is demand for new types of sensors capable of long-term operation in oxidizing environments at temperatures over 1600 ℃ to enable accurate temperature measurements in harsh environments, such as those of aero-engine turbine blades and the interiors of rocket engines.

    Ultrasonic temperature measurement technology is a novel method based on the propagation of ultrasonic waves through a medium, where the speed of sound changes with the environmental temperature. This relationship between sound speed and temperature allows for accurate temperature measurement to nearly the melting point of the material [12]. In recent years, magnesium aluminate spinel (MgAl2O4, referred to hereafter simply as "spinel") has been widely used in high-temperature furnace windows and high-temperature sensor substrates. It is favored for its excellent thermal, mechanical, and optical properties. The development of laser-heated pedestal growth (LHPG) technology for growing single crystals [13,14,15] further allows for the development of spinel ultrasonic waveguides with high length-to-diameter ratios at lengths greater than 100 mm and diameters below 1 mm. Thus, the erosion resistance, oxidation resistance, and high-temperature stability of spinel can be leveraged to take long-term, real-time temperature measurements in oxidative environments above 1600 ℃.

    Ultrasound has good directionality and anti-interference properties, making it highly suitable for measuring temperatures in challenging environments. Applying ultrasonic guided waves enhances both the measurement range and the resistance to electromagnetic interference under oxidative conditions [16,17].

    When ultrasound propagates through materials, its speed c has a certain correlation with temperature T [18,19,20]. In solid media, the longitudinal and transverse wave speeds of ultrasound can be expressed as follows:

    vL=E(1ρ)ρ(1+σ)(12σ) (1)
    vS=E2ρ(1+σ) (2)

    where vL represents the longitudinal wave speed of ultrasound, vS is the transverse wave speed of ultrasound, E is the elastic modulus of the selected waveguide material, ρ is the density of the material, and σ is Poisson's ratio. Typically, the relationship between the longitudinal wave speed of ultrasound and temperature is utilized for temperature measurement; accordingly, this equation can be transformed into Eq (3).

    v(T)=E(T)ρ(T) (3)

    The speed of ultrasound can be calculated by measuring its propagation time t and determining the gauge length l of the sensitive region, thus revealing the relationship between the speed of sound v(T) and the ambient temperature T, as expressed in Eq (4).

    Δt(T)=2lv(T) (4)

    The speed of the sound corresponding to the temperature can be determined by measuring the propagation time of the ultrasound. In the proposed equipment setup, an upper computer sends an excitation signal to an ultrasonic excitation power supply, which generates a pulse excitation. This excites an acoustic signal at an ultrasonic probe coupled to an ultrasonic waveguide. As the ultrasound encounters the prepared sensitive structure, the reflected acoustic signal reaches the probe and is converted into an electrical signal. A data acquisition system transmits this amplified signal to the upper computer for processing, calculation, and display, thereby achieving the measurement of ultrasound propagation time and determining the speed of sound corresponding to the temperature. The testing principle is illustrated in Figure 1.

    Figure 1.  Test schematic diagram.

    A key parameter of the sensor's sensitive element is the reflection coefficient, which is related to the impedance matching of the acoustic wave transmission. Acoustic wave reflection occurs when transmitted ultrasonic waves encounter a change in the diameter of the waveguide rod, a phenomenon closely related to changes in acoustic impedance [21]. The acoustic impedance at the location where the acoustic wave reflection occurs can be expressed as follows:

    Z=ρcA (5)

    and the formulas for the reflection coefficient R and transmission coefficient T at the sensitive element section are:

    R=Z2Z1Z2+Z1 (6)
    T=2Z2Z2+Z1 (7)

    where Z1 and Z2 represent the impedances before and after the change in diameter, respectively. The relationship between Z1 and Z2 is expressed in Eq (8):

    Z1Z2=d21d22 (8)

    where d1 is the initial diameter of the sensitive element (i.e., diameter of the waveguide rod) and d2 is the diameter after the change at that section. By substitution into Eq (8), R and T can be further expressed as:

    R=d22d21d22+d21 (9)
    T=2d22d22+d21 (10)

    Similarly, the diameter ratio at the variable cross-section can be obtained from the reflection coefficient T of the sensitive element:

    d2d1=1+R1R (11)

    Selecting a reasonable reflection coefficient is essential for constructing an ideal sensitive structure that combines a high signal-to-noise ratio, high sensitivity, and robust interference resistance. To minimize ultrasound dispersion and ensure the structural integrity of the sensor's waveguide rod, a spinel waveguide material with a diameter of 0.7 mm and a groove diameter of 0.57 mm was chosen for the proposed design. This effectively improves the signal-to-noise ratio, reduces the difficulty of signal acquisition and processing, and enhances measurement accuracy. The section length was designed as 25 mm according to Eq (4), balancing the necessity for a minimal section length with the difficulty of signal acquisition and processing.

    The diameter of the sensor's waveguide rod, the length and diameter of the sensitive element section, and other parameters (e.g., width) were determined by theoretical analysis. To validate these theoretical values and refine the sensor's design, finite element software was employed to simulate the ultrasound transmission within the waveguide rod; this helped in visualizing the ultrasound transmission characteristic curves and optimizing the structure of the sensor. A spinel simulation structure was constructed for this purpose. Its parameters are listed in Table 1.

    Table 1.  Sensor simulation parameters.
    Length Diameter Section length Groove diameter Reflection coefficient
    300 mm 0.7 mm 25 mm 0.57 mm 0.2

     | Show Table
    DownLoad: CSV

    The simulation results, as shown in Figure 2, indicated clear echoes at specific sections and the end of the waveguide rod, including secondary echoes. These confirm the viability of the proposed method for ultrasound signal detection. The section diameter was subsequently adjusted to 0.4 mm for additional simulations to explore the impact of different groove diameters on sensor signals. There was a significant increase in the amplitude of the echo signal at the section position, albeit with a slight decrease at the end wave echo. While the enhanced section echo signal is beneficial, the increased groove depth reduces the overall resistance to thermal shock and vibration, which would compromise its long-term durability. When the groove depth was designed to be 0.65mm, it will have a greater amplitude in the echo signal, and the amplitude of the end wave decreases. Thus, the groove depth must be carefully designed to balance signal clarity with structural integrity.

    Figure 2.  Simulation of ultrasonic propagation characteristics: (a)the groove diameter is 0.57 mm, (b) the groove diameter is 0.4 mm, and (c) the groove diameter is 0.65 mm.

    Additional simulations were conducted to examine the ultrasound propagation characteristics across different temperatures, producing a relationship curve between the speed of sound and temperature (Figure 3). The speed of sound in the spinel gradually decreased as the temperature increased, from 9261 m/s at room temperature to 8569 m/s at 1600℃, demonstrating a clear monotonous decreasing trend that corroborates the theoretical predictions.

    Figure 3.  Simulated relationship between temperature and speed during ultrasonic propagation.

    In recent years, significant progress has been made in the research of functional crystal materials. These materials are becoming more widely used in the sensor industry as main materials, substrates, and infrared window components. Spinel materials, which have melting points as high as 2150 ℃, possess excellent thermal, mechanical, optical, and electrical properties.

    Spinel performs similarly to the more commonly used sapphire, with properties including a thermal expansion coefficient of 7.33×10-6/ ℃, a melting point of 2135 ℃, and a Mohs hardness of 8.5. In comparison, alumina has a melting point of 2050 ℃, a Mohs hardness of 9, and a thermal expansion coefficient of 7.5×10–6/ ℃. However, spinel has superior chemical stability and, due to its cubic structure and optical isotropy, is an ideal material for transparent substrates and sublayers. These performance characteristics are outlined in Table 2.

    Table 2.  Spinel versus sapphire.
    Category Molecular formula Melting point
    (℃)
    Elastic modulus
    (GPa)
    Poisson's ratio Density
    (g/cm3)
    Spinel MgAl2O4 2135 273 0.26 3.58
    Sapphire α-Al2O3 2050 344 0.23 3.97

     | Show Table
    DownLoad: CSV

    The LHPG method is extensively used to grow high-quality single-crystal fibers, particularly in producing ultrasonic waveguides. Control over the pulling speed of the seed crystal during the growth process is crucial for successfully fabricating such waveguides. This parameter significantly influences both ultrasonic transmission and the sensitivity of sensing elements to environmental temperatures.

    To operate the proposed method, a focused laser is used to heat the source rod to create a molten zone into which the seed crystal is inserted. The geometric parameters of the single-crystal ultrasonic waveguide are regulated by adjusting the pulling speed of the seed crystal vc and the rising speed of the source rod vr, allowing the growth of single-crystal waveguides with specific diameters tailored to the needs of the application at hand. A schematic diagram illustrating this growth apparatus is shown in Figure 4.

    Figure 4.  Single crystal fiber growth device.

    To grow an ultrasonic waveguide with a uniform shape and no undulations, the pulling speed of the seed crystal vc and the rising speed of the source rod vr should satisfy the following relationship:

    vcvr=(DcDr)2 (12)

    where Dc represents the diameter of the seed crystal and Dr is the diameter of the source rod.

    In this study, a high aspect ratio ultrasonic waveguide single-crystal fiber with a length of 300 mm and a diameter of 0.7 mm was grown using a spinel waveguide seed crystal, as illustrated in Figure 5.

    Figure 5.  Growth of spinel ultrasonic waveguide.

    An innovative system was constructed in this study based on the principle of ultrasonic temperature measurement. The system consists of a temperature-sensitive element, an ultrasonic transmitting and receiving device, an ultrasonic signal acquisition and processing system, and a probe coupling device, as depicted in Figure 6.

    Figure 6.  An ultrasonic temperature measuring system.

    The temperature-sensitive element was manufactured using precision engraving methods to form a temperature-sensitive structure. A CTS-8077PR pulse transmitter-receiver was used for the transmission and reception of ultrasound, employing a 2.5 MHz ultrasonic probe. This setup facilitated the accurate capture of echo signals, which were then processed to extract characteristic wave signals to compute the ultrasonic propagation time. The system was evaluated using a high-temperature resistance furnace capable of providing a controlled temperature range from 20℃ to 1600℃. Readings were taken at 100℃ intervals throughout the testing process, with each point maintained for 10 min to ensure equilibrium between the sensor temperature and ambient conditions before proceeding with signal acquisition. A standard platinum-rhodium thermocouple was used as the reference temperature standard for calibration purposes.

    The ultrasonic signals were initially processed to remove the DC component. After conducting multiple experiments, a sliding window length of 9 was found to yield the optimal filtering effect. The echo signal's spectrum exhibited bandpass characteristics. The passband center frequency was set to 2.5 MHz to match the center frequency of the ultrasonic transducer. For enhanced computational efficiency and speed, convolution operations were executed using both FFT and IFFT algorithms. The relationship between the node wave and the endpoint wave was established using their cross-correlation function based on their correlation and time difference [22]. The time delay was identified by the peak value in the cross-correlation curve, which reflects the delay between the two signals. The ultrasonic delay time was then calculated by dividing the number of sampling points from the starting point to this peak by the sampling frequency. Since the node wave and end point are positively correlated, a correlation coefficient of 1 was assigned. This algorithm boasts high accuracy, rendering it well-suited for ultrasonic signal processing; however, it incurs substantial computational demands and requires extended computation times. To perform fast calculations on the upper computer, the algorithm has been simplified to reduce computational complexity and speed up the computation. However, when measuring continuously, if the amount of data is too large, the calculation speed is a bit slow. A flowchart of this algorithm is presented in Figure 7, and the parameters of the algorithm are listed in Table 3.

    Figure 7.  Algorithm flowchart.
    Table 3.  Algorithm parameters.
    Parameters Sliding window length Center frequency Correlation coefficient
    Value 9 2.5MHz 1

     | Show Table
    DownLoad: CSV

    The LHPG method was used to prepare spinel and magnesium-doped alumina ultrasonic waveguides. Ultrasonic temperature sensors were created from these materials and each was subjected to three round-trip tests. The resultant data were analyzed using specialized algorithms to determine the ultrasonic propagation time, allowing for calculating the relationship between ultrasonic propagation temperature and velocity. The average sensitivity and repeatability of the sensors were also evaluated.

    For the spinel ultrasonic sensor, three cycles of heating and cooling tests revealed temperature-velocity and temperature-delay time relationship curves within the temperature range of 20 ℃ to 1600 ℃, as shown in Figure 8. As the temperature increased during the forward tests, the ultrasonic propagation speed in the spinel decreased from 9250 m/s to 8500 m/s, and the propagation time increased from 5.4 μs to 5.85 μs (Figure 8(a), Figure 8(b)). During the backward tests, as the temperature decreased, the ultrasonic propagation speed in the spinel increased from 8500 m/s to 9250 m/s and the propagation time recovered from 5.85 μs to 5.4 μs (Figure 8(c), Figure 8(d)). These trends in velocity and propagation time are consistent with the simulation results.

    Figure 8.  Temperature-sound velocity curve of self-made ultrasonic waveguide.

    Based on Formulas (13) and (14), the sensitivity of the sensor was calculated to be 0.48 m/s·℃ with a repeatability of 95%.

    Average sensitivity:

    s=YmaxYminXmaxXmin (13)

    Repeatability:

    ξR=cSmaxYFS×100% (14)

    Similar forward and backward tests were also conducted for the magnesium-doped alumina ultrasonic sensor. Temperature-velocity and temperature-delay time relationship curves for the sensor were obtained within the temperature range of 50℃ to 1600℃, as shown in Figure 9. During the forward tests, as the temperature increased, the ultrasonic propagation speed in the spinel decreased from 10505 m/s to 9551 m/s, and the propagation time increased from 5.15 μs to 5.65 μs (Figure 9(a), Figure 9(b)). As the temperature decreased during the backward tests, the ultrasonic propagation speed in the spinel increased from 9551 m/s to 10505 m/s, and the propagation time recovered from 5.65 μs to 5.15 μs (Figure 9(c), Figure 9(d)). These trends in velocity and propagation time are also consistent with the simulation results. Based on Formula (13, 14), the sensitivity of the sensor was calculated to be 0.615m/s·℃, with a repeatability of 97%.

    Figure 9.  Magnesium-doped alumina.

    As per comparisons of the transmission characteristics of ultrasonic guided waves in spinel and magnesium-doped alumina, the magnesium-doped alumina waveguide exhibited less noise in the acoustic signal compared to the spinel material under identical filtering conditions. At a temperature of 1600℃, the amplitude of the spinel acoustic waveguide remained almost unchanged compared to that at 50℃, while the amplitude of the magnesium-doped alumina acoustic waveguide decreased by approximately 70%. However, at high temperatures, the spinel waveguide produced more distinguishable signal waveforms. The magnesium doping in alumina introduced instability that increased the acoustic impedance within the crystal at elevated temperatures, resulting in a noticeable decrease in signal waveform amplitude at higher temperatures compared to lower ones. This complicated the extraction of temperature signals, as depicted in Figure 10. Researchers should focus on enhancing the stability of magnesium doping in these applications in the future.

    Figure 10.  Ultrasonic characteristics of spinel and magnesia-doped alumina at different temperatures: (a) Spinel and (b) Magnesia-doped alumina.

    Tests conducted within the temperature range of 20 ℃ to 1600 ℃ for three types of crystal ultrasonic waveguides (sapphire, magnesium-doped alumina, and spinel) revealed variations in ultrasonic velocities at different temperatures, as shown in Figure 11. From 50℃ to 1600℃, the ultrasonic propagation speed of spinel decreased from 9271 m/s to 8527 m/s, the ultrasonic propagation speed of magnesium-doped alumina decreased from 10505 m/s to 9551 m/s, and the sound velocity of sapphire decreased from 10666 m/s to 10000 m/s.

    Figure 11.  Spinel (Plot 1); Magnesium-doped sapphire (Plot 2); and Sapphire (Plot 3).

    The overall trend indicates a general decrease in sound velocity with increasing temperature. Sapphire exhibited the highest velocity, followed by magnesium-doped alumina, while spinel showed the lowest velocity at equivalent temperatures. While the difference in sound velocity between magnesium-doped alumina and pure alumina was initially slight, it became more pronounced as temperature increased. At the same temperature and acquisition frequency, a lower sound velocity can help to reduce the length of the cell segment, thus facilitating sensor miniaturization.

    Temperature measurement in challenging environments, such as in aviation and aerospace engines, is a pressing issue that requires innovative solutions. Based on the principle of ultrasonic temperature sensing, structural parameters for a magnesium-aluminum spinel ultrasonic temperature sensor were designed in this study. The proposed design includes a waveguide diameter of 0.7 mm, a section length of 25 mm, and a groove diameter of 0.57 mm. Simulations of ultrasonic propagation characteristics at different temperatures revealed a decreasing trend in sound velocity with increasing temperature. A magnesium-aluminum spinel ultrasonic waveguide was grown using the LHPG method, then an ultrasonic temperature sensor was fabricated according to the designed parameters. Calibration results demonstrated a sensitivity of 0.48 m/s·℃ and repeatability of 95% within the temperature range of 20 ℃ to 1600 ℃.

    Comparative analysis of ultrasonic characteristics among magnesium-doped alumina, single-crystal alumina, and magnesium-aluminum spinel revealed a gradual decrease in sound velocities. At the same temperature, single-crystal alumina exhibited the highest sound velocity, followed by magnesium-doped alumina, while magnesium-aluminum spinel had the lowest. Utilizing magnesium-aluminum spinel as an ultrasonic waveguide material can effectively reduce sound velocity, improve the sensor's resolution, and simplify the processing of ultrasonic signals. By leveraging ultrasonic temperature sensing combined with the oxidation resistance of oxide single-crystal materials, there is significant potential to develop an effective solution for high-temperature measurement in the highly oxidizing and highly abrasive environments typical of aviation and aerospace engines.

    H. J. Liang: systems design, conceptualization, simulation analysis, data curation, writing-original draft; X. H. Wang: data curation, experimental design, writing-review and editing; H. X. Xue: algorithm design, writing-review and editing. All authors have read and agreed to the published version of the article.

    The authors declare they have not used Artificial Intelligence (AI) tools in the creation of this article.

    This paper is supported by the National Natural Science Foundation of China granted [No. 62403440, 62106238], Aeronautical Science Foundation of China [No. 202300340U0002], Fundamental Research Program of Shanxi Province [No. 20210302124541].

    All authors declare no conflicts of interest in this paper.



    [1] M. E. Sahin, Image processing and machine learning‐based bone fracture detection and classification using X‐ray images, Int. J. Imaging Syst. Technol., 33 (2023), 853–865. https://doi.org/10.1002/ima.22849 doi: 10.1002/ima.22849
    [2] Z. Zhao, Y. Tian, Z. Yuan, P. Zhao, F. Xia, S. Yu, A machine learning method for improving liver cancer staging, J. Biomed. Inf., 137 (2023), 104266. https://doi.org/10.1002/ima.22849 doi: 10.1002/ima.22849
    [3] S. Maurya, S. Tiwari, M. C. Mothukuri, C. M. Tangeda, R. N. S. Nandigam, D. C. Addagiri, A review on recent developments in cancer detection using Machine Learning and Deep Learning models, Biomed. Signal Process. Control, 80 (2023), 104398. https://doi.org/10.1016/j.bspc.2022.104398 doi: 10.1016/j.bspc.2022.104398
    [4] A. Radford, K. Narasimhan, T. Salimans, I. Sutskever, Improving language understanding by generative pre-training, OpenAI, 2018.
    [5] A. Ramesh, P. Dhariwal, A. Nichol, C. Chu, M. Chen, Hierarchical text-conditional image generation with CLIP latents, preprint, arXiv.2204.06125. https://doi.org/10.48550/arXiv.2204.06125
    [6] A. J. Thirunavukarasu, D. S. J. Ting, K. Elangovan, L. Gutierrez, T. F. Tan, D. S. W. Ting, Large language models in medicine, Nat. Med., 29 (2023), 1930–1940. https://doi.org/10.1038/s41591-023-02448-8 doi: 10.1038/s41591-023-02448-8
    [7] A. Radford, J. Wu, R. Child, D. Luan, D. Amodei, I. Sutskever, Language models are unsupervised multitask learners, OpenAI blog, 1 (2019), 9.
    [8] T. Brown, B. Mann, N. Ryder, M. Subbiah, J. D. Kaplan, P. Dhariwal, et al., Language models are few-shot learners, Adv. Neural Inf. Process. Syst., 33 (2020), 1877–1901.
    [9] S. Bubeck, V. Chandrasekaran, R. Eldan, J. Gehrke, E. Horvitz, E. Kamar, et al., Sparks of artificial general intelligence: Early experiments with gpt-4, preprint, arXiv: 2303.12712. https://doi.org/10.48550/arXiv.2303.12712
    [10] J. W. Rae, S. Borgeaud, T. Cai, K. Millican, J. Hoffmann, F. Song, et al., Scaling language models: Methods, analysis & insights from training gopher, preprint, arXiv: 2112.11446. https://doi.org/10.48550/arXiv.2112.11446
    [11] T. L. Scao, A. Fan, C. Akiki, E. Pavlick, S. Ilić, D. Hesslow, et al., Bloom: A 176b-parameter open-access multilingual language model, preprint, arXiv: 2211.05100. https://doi.org/10.48550/arXiv.2211.05100
    [12] L. Ouyang, J. Wu, X. Jiang, D. Almeida, C. Wainwright, P. Mishkin, et al., Training language models to follow instructions with human feedback, Adv. Neural Inf. Process. Syst., 35 (2022), 27730–27744.
    [13] C. Raffel, N. Shazeer, A. Roberts, K. Lee, S. Narang, M. Matena, et al., Exploring the limits of transfer learning with a unified text-to-text transformer, J. Machine Learn. Res., 21 (2020), 5485–5551.
    [14] W. Fedus, B. Zoph, N. Shazeer, Switch transformers: Scaling to trillion parameter models with simple and efficient sparsity, J. Machine Learn. Res., 23 (2022), 5232–5270.
    [15] V. Aribandi, Y. Tay, T. Schuster, J. Rao, H. S. Zheng, S. V. Mehta, et al., Ext5: Towards extreme multi-task scaling for transfer learning, preprint, arXiv: 2111.10952. https://doi.org/10.48550/arXiv.2111.10952
    [16] M. Lewis, Y. Liu, N. Goyal, M. Ghazvininejad, A. Mohamed, O. Levy, et al., Bart: Denoising sequence-to-sequence pre-training for natural language generation, translation, and comprehension, preprint, arXiv: 1910.13461. https://doi.org/10.48550/arXiv.1910.13461
    [17] Z. Li, Z. Wang, M. Tan, R. Nallapati, P. Bhatia, A. Arnold, et al., Dq-bart: Efficient sequence-to-sequence model via joint distillation and quantization, preprint, arXiv: 2203.11239. https://doi.org/10.48550/arXiv.2203.11239
    [18] I. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde-Farley, S. Ozair, et al., Generative adversarial networks, Commun. ACM, 63 (2020), 139–144. https://doi.org/10.1145/3422622 doi: 10.1145/3422622
    [19] D. P. Kingma, M. Welling, Auto-encoding variational bayes, preprint, arXiv: 1312.6114. https://doi.org/10.48550/arXiv.1312.6114
    [20] L. Dinh, D. Krueger, Y. Bengio, Nice: Non-linear independent components estimation, preprint, arXiv: 1410.8516. https://doi.org/10.48550/arXiv.1410.8516
    [21] Y. Song, S. Ermon, Generative modeling by estimating gradients of the data distribution, Adv. Neural Inf. Process. Syst., 32 (2019).
    [22] E. L. Denton, S. Chintala, R. Fergus, Deep generative image models using a laplacian pyramid of adversarial networks, Adv. Neural Inf. Process. Syst., 28 (2015).
    [23] H. Zhang, I. Goodfellow, D. Metaxas, A. Odena, Self-attention generative adversarial networks, in International Conference on Machine Learning, (2019), 7354–7363.
    [24] A. Radford, L. Metz, S. Chintala, Unsupervised representation learning with deep convolutional generative adversarial networks, preprint, arXiv: 1511.06434. https://doi.org/10.48550/arXiv.1511.06434
    [25] M. Liu, O. Tuzel, Coupled generative adversarial networks, Adv. Neural Inf. Process. Syst., 29 (2016).
    [26] A. Brock, J. Donahue, K. Simonyan, Large scale GAN training for high fidelity natural image synthesis, preprint, arXiv: 1809.11096. https://doi.org/10.48550/arXiv.1809.11096
    [27] T. Nguyen, T. Le, H. Vu, D. Phung, Dual discriminator generative adversarial nets, Adv. Neural Inf. Process. Syst., 30 (2017).
    [28] I. Durugkar, I. Gemp, S. Mahadevan, Generative multi-adversarial networks, preprint, arXiv: 1611.01673. https://doi.org/10.48550/arXiv.1611.01673
    [29] Q. Hoang, T. D. Nguyen, T. Le, D. Phung, Multi-generator generative adversarial nets, preprint, arXiv: 1708.02556. https://doi.org/10.48550/arXiv.1708.02556
    [30] A. Ghosh, V. Kulharia, V. P. Namboodiri, P. H. Torr, P. K. Dokania, Multi-agent diverse generative adversarial networks, in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, (2018), 8513–8521. https://doi.org/10.1109/CVPR.2018.00888
    [31] S. Nowozin, B. Cseke, R. Tomioka, f-gan: Training generative neural samplers using variational divergence minimization, Adv. Neural Inf. Process. Syst., 29 (2016).
    [32] T. Miyato, T. Kataoka, M. Koyama, Y. Yoshida, Spectral normalization for generative adversarial networks, preprint, arXiv: 1802.05957. https://doi.org/10.48550/arXiv.1802.05957
    [33] I. Gulrajani, F. Ahmed, M. Arjovsky, V. Dumoulin, A. C. Courville, Improved training of wasserstein gans, Adv. Neural Inf. Process. Syst., 30 (2017).
    [34] G. Qi, Loss-sensitive generative adversarial networks on lipschitz densities, Int. J. Comput. Vis., 128 (2020), 1118–1140. https://doi.org/10.1007/s11263-019-01265-2 doi: 10.1007/s11263-019-01265-2
    [35] T. Che, Y. Li, A. P. Jacob, Y. Bengio, W. Li, Mode regularized generative adversarial networks, preprint, arXiv: 1612.02136. https://doi.org/10.48550/arXiv.1612.02136
    [36] L. Maaløe, M. Fraccaro, V. Liévin, O. Winther, Biva: A very deep hierarchy of latent variables for generative modeling, Adv. Neural Inf. Process. Syst., 32 (2019).
    [37] A. Vahdat, J. Kautz, NVAE: A deep hierarchical variational autoencoder, Adv. Neural Inf. Process. Syst., 33 (2020), 19667–19679.
    [38] B. Wu, S. Nair, R. Martin-Martin, L. Fei-Fei, C. Finn, Greedy hierarchical variational autoencoders for large-scale video prediction, in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, (2021), 2318–2328.
    [39] A. Ramesh, M. Pavlov, G. Goh, S. Gray, C. Voss, A. Radford, et al., Zero-shot text-to-image generation, in International Conference on Machine Learning, (2021), 8821–8831.
    [40] P. Ghosh, M. S. Sajjadi, A. Vergari, M. Black, B. Schölkopf, From variational to deterministic autoencoders, preprint, arXiv: 1903.12436. https://doi.org/10.48550/arXiv.1903.12436
    [41] A. V. D. Oord, O. Vinyals, Neural discrete representation learning, Adv. Neural Inf. Process. Syst., 30 (2017).
    [42] A. Razavi, A. V. Oord, O. Vinyals, Generating diverse high-fidelity images with vq-vae-2, Adv. Neural Inf. Process. Syst., 32 (2019).
    [43] G. Zheng, Y. Yang, J. Carbonell, Convolutional normalizing flows, preprint, arXiv: 1711.02255. https://doi.org/10.48550/arXiv.1711.02255
    [44] E. Hoogeboom, R. Van Den Berg, M. Welling, Emerging convolutions for generative normalizing flows, in International Conference on Machine Learning, (2019), 2771–2780.
    [45] A. N. Gomez, M. Ren, R. Urtasun, R. B. Grosse, The reversible residual network: Backpropagation without storing activations, Adv. Neural Inf. Process. Syst., 30 (2017).
    [46] J. Jacobsen, A. Smeulders, E. Oyallon, i-revnet: Deep invertible networks, preprint, arXiv: 1802.07088. https://doi.org/10.48550/arXiv.1802.07088
    [47] T. Salimans, J. Ho, Progressive distillation for fast sampling of diffusion models, preprint, arXiv: 2202.00512. https://doi.org/10.48550/arXiv.2202.00512
    [48] E. Luhman, T. Luhman, Knowledge distillation in iterative generative models for improved sampling speed, preprint, arXiv: 2101.02388. https://doi.org/10.48550/arXiv.2101.02388
    [49] Z. Kong, W. Ping, On fast sampling of diffusion probabilistic models, preprint, arXiv: 2106.00132. https://doi.org/10.48550/arXiv.2106.00132
    [50] A. Q. Nichol, P. Dhariwal, Improved denoising diffusion probabilistic models, in International Conference on Machine Learning, (2021), 8162–8171.
    [51] D. Kingma, T. Salimans, B. Poole, J. Ho, Variational diffusion models, Adv. Neural Inf. Process. Syst., 34 (2021), 21696–21707.
    [52] R. San-Roman, E. Nachmani, L. Wolf, Noise estimation for generative diffusion models, preprint, arXiv: 2104.02600. https://doi.org/10.48550/arXiv.2104.02600
    [53] D. Watson, W. Chan, J. Ho, M. Norouzi, Learning fast samplers for diffusion models by differentiating through sample quality, in International Conference on Learning Representations, 2021.
    [54] D. Watson, J. Ho, M. Norouzi, W. Chan, Learning to efficiently sample from diffusion probabilistic models, preprint, arXiv: 2106.03802. https://doi.org/10.48550/arXiv.2106.03802
    [55] H. Zheng, P. He, W. Chen, M. Zhou, Truncated diffusion probabilistic models, preprint, arXiv: 2202.09671. https://doi.org/10.48550/arXiv.2202.09671
    [56] K. Pandey, A. Mukherjee, P. Rai, A. Kumar, Diffusevae: Efficient, controllable and high-fidelity generation from low-dimensional latents, preprint, arXiv: 2201.00308. https://doi.org/10.48550/arXiv.2201.00308
    [57] Q. Zhang, Y. Chen, Diffusion normalizing flow, Adv. Neural Inf. Process. Syst., 34 (2021), 16280–16291.
    [58] L. H. Li, M. Yatskar, D. Yin, C. Hsieh, K. Chang, Visualbert: A simple and performant baseline for vision and language, preprint, arXiv: 1908.03557. https://doi.org/10.48550/arXiv.1908.03557
    [59] L. Zhou, H. Palangi, L. Zhang, H. Hu, J. Corso, J. Gao, Unified vision-language pre-training for image captioning and vqa, in Proceedings of the AAAI Conference on Artificial Intelligence, (2020), 13041–13049.
    [60] H. Tan, M. Bansal, Lxmert: Learning cross-modality encoder representations from transformers, preprint, arXiv: 1908.07490. https://doi.org/10.48550/arXiv.1908.07490
    [61] J. Lu, D. Batra, D. Parikh, S. Lee, Vilbert: Pretraining task-agnostic visiolinguistic representations for vision-and-language tasks, Adv. Neural Inf. Process. Syst., 32 (2019).
    [62] M. Tsimpoukelli, J. L. Menick, S. Cabi, S. M. Eslami, O. Vinyals, F. Hill, Multimodal few-shot learning with frozen language models, Adv. Neural Inf. Process. Syst., 34 (2021), 200–212.
    [63] O. Patashnik, Z. Wu, E. Shechtman, D. Cohen-Or, D. Lischinski, Styleclip: Text-driven manipulation of stylegan imagery, in Proceedings of the IEEE/CVF International Conference on Computer Vision, (2021), 2085–2094. https://doi.org/10.1109/ICCV48922.2021.00209
    [64] A. Nichol, P. Dhariwal, A. Ramesh, P. Shyam, P. Mishkin, B. McGrew, et al., Glide: Towards photorealistic image generation and editing with text-guided diffusion models, preprint, arXiv: 2112.10741. https://doi.org/10.48550/arXiv.2112.10741
    [65] C. Saharia, W. Chan, S. Saxena, L. Li, J. Whang, E. L. Denton, et al., Photorealistic text-to-image diffusion models with deep language understanding, Adv. Neural Inf. Process. Syst., 35 (2022), 36479–36494. https://doi.org/10.1145/3528233.3530757 doi: 10.1145/3528233.3530757
    [66] M. Chen, X. Tan, B. Li, Y. Liu, T. Qin, S. Zhao, et al., Adaspeech: Adaptive text to speech for custom voice, preprint, arXiv: 2103.00993. https://doi.org/10.48550/arXiv.2103.00993
    [67] H. Liang, H. Wang, J. Wang, S. You, Z. Sun, J. Wei, et al., JTAV: Jointly learning social media content representation by fusing textual, acoustic, and visual features, preprint, arXiv: 1806.01483. https://doi.org/10.48550/arXiv.1806.01483
    [68] Z. Feng, D. Guo, D. Tang, N. Duan, X. Feng, M. Gong, et al., Codebert: A pre-trained model for programming and natural languages, preprint, arXiv: 2002.08155. https://doi.org/10.48550/arXiv.2002.08155
    [69] W. U. Ahmad, S. Chakraborty, B. Ray, K. Chang, Unified pre-training for program understanding and generation, preprint, arXiv: 2103.06333. https://doi.org/10.48550/arXiv.2103.06333
    [70] I. Melnyk, P. Dognin, P. Das, Knowledge graph generation from text, preprint, arXiv: 2211.10511. https://doi.org/10.48550/arXiv.2211.10511
    [71] B. Distiawan, J. Qi, R. Zhang, W. Wang, GTR-LSTM: A triple encoder for sentence generation from RDF data, in Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, 1 (2018), 1627–1637.
    [72] M. Li, J. Wang, Y. Chen, Y. Tang, Z. Wu, Y. Qi, et al., Low-dose CT image synthesis for domain adaptation imaging using a generative adversarial network with noise encoding transfer learning, IEEE Trans. Med. Imaging, 2023.
    [73] Q. Gao, Z. Li, J. Zhang, Y. Zhang, H. Shan, CoreDiff: Contextual Error-Modulated Generalized Diffusion Model for Low-Dose CT Denoising and Generalization, preprint, arXiv: 2304.01814. https://doi.org/10.48550/arXiv.2304.01814
    [74] Z. Huang, J. Zhang, Y. Zhang, H. Shan, DU-GAN: Generative adversarial networks with dual-domain U-Net-based discriminators for low-dose CT denoising, IEEE Trans. Instrum. Meas., 71 (2021), 1–12. https://doi.org/10.1109/TIM.2021.3128703 doi: 10.1109/TIM.2021.3128703
    [75] B. Chen, S. Leng, L. Yu, D. Holmes III, J. Fletcher, C. McCollough, An open library of CT patient projection data, in Medical Imaging 2016: Physics of Medical Imaging, 9783 (2016), 330–335. https://doi.org/10.1117/12.2216823
    [76] X. Zhao, T. Yang, B. Li, X. Zhang, SwinGAN: A dual-domain Swin Transformer-based generative adversarial network for MRI reconstruction, Comput. Biol. Med., 153 (2023), 106513. https://doi.org/10.1016/j.compbiomed.2022.106513 doi: 10.1016/j.compbiomed.2022.106513
    [77] C. Zhang, R. Barbano, B. Jin, Conditional variational autoencoder for learned image reconstruction, Computation, 9 (2021), 114. https://doi.org/10.3390/computation9110114 doi: 10.3390/computation9110114
    [78] G. Luo, M. Heide, M. Uecker, MRI reconstruction via data driven markov chain with joint uncertainty estimation, preprint, arXiv: 2202.01479. https://doi.org/10.48550/arXiv.2202.01479
    [79] Y. Gu, Z. Zeng, H. Chen, J. Wei, Y. Zhang, B. Chen, et al., MedSRGAN: medical images super-resolution using generative adversarial networks, Multimed. Tools Appl., 79 (2020), 21815–21840. https://doi.org/10.1007/s11042-020-08980-w doi: 10.1007/s11042-020-08980-w
    [80] A. A. A. Setio, A. Traverso, T. D. Bel, M. S. Berens, C. V. D. Bogaard, P. Cerello, et al., Validation, comparison, and combination of algorithms for automatic detection of pulmonary nodules in computed tomography images: the LUNA16 challenge, Med. Image Anal., 42 (2017), 1–13. https://doi.org/10.1016/j.media.2017.06.015 doi: 10.1016/j.media.2017.06.015
    [81] B. Vasudeva, P. Deora, S. Bhattacharya, P. M. Pradhan, Co-VeGAN: Complex-valued generative adversarial network for compressive sensing MR image reconstruction, preprint, arXiv: 2002.10523. https://doi.org/10.48550/arXiv.2002.10523
    [82] B. Landman, S. Warfield, Diencephalon standard challenge, 2013. https://doi.org/10.7303/syn3270351
    [83] N. Bien, P. Rajpurkar, R. L. Ball, J. Irvin, A. Park, E. Jones, et al., Deep-learning-assisted diagnosis for knee magnetic resonance imaging: development and retrospective validation of MRNet, PLoS Med., 15 (2018), e1002699. https://doi.org/10.1371/journal.pmed.1002699 doi: 10.1371/journal.pmed.1002699
    [84] J. Zbontar, F. Knoll, A. Sriram, T. Murrell, Z. Huang, M. J. Muckley, et al., fastMRI: An open dataset and benchmarks for accelerated MRI, preprint, arXiv: 1811.08839. https://doi.org/10.48550/arXiv.1811.08839
    [85] Z. Yuan, M. Jiang, Y. Wang, B. Wei, Y. Li, P. Wang, et al., SARA-GAN: Self-attention and relative average discriminator based generative adversarial networks for fast compressed sensing MRI reconstruction, Front. Neuroinf., 14 (2020), 611666. https://doi.org/10.3389/fninf.2020.611666 doi: 10.3389/fninf.2020.611666
    [86] M. Zehni, Z. Zhao, UVTOMO-GAN: An adversarial learning based approach for unknown view X-ray tomographic reconstruction, in 2021 IEEE 18th International Symposium on Biomedical Imaging (ISBI), (2021), 1812–1816. https://doi.org/10.1109/ISBI48211.2021.9433970
    [87] B. Gajera, S. R. Kapil, D. Ziaei, J. Mangalagiri, E. Siegel, D. Chapman, CT-scan denoising using a charbonnier loss generative adversarial network, IEEE Access, 9 (2021), 84093–84109. https://doi.org/10.1109/ACCESS.2021.3087424 doi: 10.1109/ACCESS.2021.3087424
    [88] M. A. Gavrielides, L. M. Kinnard, K. J. Myers, J. Peregoy, W. F. Pritchard, R. Zeng, et al., Data from phantom FDA. The cancer imaging archive, Nat. Cancer Inst., Bethesda, MD, USA, Tech. Rep, 2015.
    [89] A. Aghabiglou, E. M. Eksioglu, MR image reconstruction based on densely connected residual generative adversarial network–DCR-GAN, in Advances in Computational Collective Intelligence: 13th International Conference, ICCCI 2021, Kallithea, Rhodes, Greece, September 29–October 1, 2021, Proceedings 13, (2021), 679–689. https://doi.org/10.1007/978-3-030-88113-9_55
    [90] J. Lv, C. Wang, G. Yang, PIC-GAN: a parallel imaging coupled generative adversarial network for accelerated multi-channel MRI reconstruction, Diagnostics, 11 (2021), 61. https://doi.org/10.3390/diagnostics11010061 doi: 10.3390/diagnostics11010061
    [91] M. Jiang, M. Zhi, L. Wei, X. Yang, J. Zhang, Y. Li, et al., FA-GAN: Fused attentive generative adversarial networks for MRI image super-resolution, Comput. Med. Imaging. Graph., 92 (2021), 101969. https://doi.org/10.1016/j.compmedimag.2021.101969 doi: 10.1016/j.compmedimag.2021.101969
    [92] S. Kyung, J. Won, S. Pak, G. Hong, N. Kim, MTD-GAN: Multi-task discriminator based generative adversarial networks for low-dose CT denoising, in International Workshop on Machine Learning for Medical Image Reconstruction, (2022), 133–144. https://doi.org/10.1007/978-3-031-17247-2_14
    [93] H. Zhou, X. Liu, H. Wang, Q. Chen, R. Wang, Z. Pang, et al., The synthesis of high-energy CT images from low-energy CT images using an improved cycle generative adversarial network, Quant. Imaging Med. Surg., 12 (2022), 28. https://doi.org/10.21037/qims-21-182 doi: 10.21037/qims-21-182
    [94] M. Yaqub, F. Jinchao, S. Ahmed, K. Arshid, M. A. Bilal, M. P. Akhter, et al., Gan-tl: Generative adversarial networks with transfer learning for mri reconstruction, Appl. Sci., 12 (2022), 8841. https://doi.org/10.3390/app12178841 doi: 10.3390/app12178841
    [95] X. Liu, H. Du, J. Xu, B. Qiu, DBGAN: A dual-branch generative adversarial network for undersampled MRI reconstruction, Magn. Reson. Imaging, 89 (2022), 77–91. https://doi.org/10.1016/j.mri.2022.03.003 doi: 10.1016/j.mri.2022.03.003
    [96] K. Zhang, H. Hu, K. Philbrick, G. M. Conte, J. D. Sobek, P. Rouzrokh, et al., SOUP-GAN: Super-resolution MRI using generative adversarial networks, Tomography, 8 (2022), 905–919. https://doi.org/10.3390/tomography8020073 doi: 10.3390/tomography8020073
    [97] H. Chung, J. C. Ye, Score-based diffusion models for accelerated MRI, Med. Image Anal., 80 (2022), 102479. https://doi.org/10.1016/j.media.2022.102479 doi: 10.1016/j.media.2022.102479
    [98] A. Güngör, S. U. Dar, Ş. Öztürk, Y. Korkmaz, H. A. Bedel, G. Elmas, et al., Adaptive diffusion priors for accelerated MRI reconstruction, Med. Image Anal., (2023), 102872. https://doi.org/10.1016/j.media.2023.102872 doi: 10.1016/j.media.2023.102872
    [99] C. Peng, P. Guo, S. K. Zhou, V. M. Patel, R. Chellappa, Towards performant and reliable undersampled MR reconstruction via diffusion model sampling, in International Conference on Medical Image Computing and Computer-Assisted Intervention, (2022), 623–633. https://doi.org/10.1007/978-3-031-16446-0_59
    [100] A. D. Desai, A. M. Schmidt, E. B. Rubin, C. M. Sandino, M. S. Black, V. Mazzoli, et al., Skm-tea: A dataset for accelerated mri reconstruction with dense image labels for quantitative clinical evaluation, preprint, arXiv: 2203.06823. https://doi.org/10.48550/arXiv.2203.06823
    [101] Y. Xie, Q. Li, Measurement-conditioned denoising diffusion probabilistic model for under-sampled medical image reconstruction, in International Conference on Medical Image Computing and Computer-Assisted Intervention, (2022), 655–664. https://doi.org/10.1007/978-3-031-16446-0_62
    [102] X. Liu, Y. Xie, S. Diao, S. Tan, X. Liang, A diffusion probabilistic prior for low-dose CT image denoising, preprint, arXiv: 2305.15887. https://doi.org/10.48550/arXiv.2305.15887
    [103] Q. Gao, H. Shan, CoCoDiff: a contextual conditional diffusion model for low-dose CT image denoising, in Developments in X-Ray Tomography XIV, 2022.
    [104] Z. Cui, C. Cao, S. Liu, Q. Zhu, J. Cheng, H. Wang, et al., Self-score: Self-supervised learning on score-based models for mri reconstruction, preprint, arXiv: 2209.00835. https://doi.org/10.48550/arXiv.2209.00835
    [105] W. Xia, Q. Lyu, G. Wang, Low-Dose CT Using Denoising Diffusion Probabilistic Model for 20× Speedup, preprint, arXiv: 2209.15136. https://doi.org/10.48550/arXiv.2209.15136
    [106] B. Huang, L. Zhang, S. Lu, B. Lin, W. Wu, Q. Liu, One sample diffusion model in projection domain for low-dose CT imaging, preprint, arXiv: 2212.03630. https://doi.org/10.48550/arXiv.2212.03630
    [107] B. Zhao, T. Cheng, X. Zhang, J. Wang, H. Zhu, R. Zhao, et al., CT synthesis from MR in the pelvic area using residual transformer conditional GAN, Comput. Med. Imaging. Graph., 103 (2023), 102150. https://doi.org/10.1016/j.compmedimag.2022.102150 doi: 10.1016/j.compmedimag.2022.102150
    [108] X. Li, K. Shang, G. Wang, M. D. Butala, DDMM-Synth: A denoising diffusion model for cross-modal medical image synthesis with sparse-view measurement embedding, preprint, arXiv: 2303.15770. https://doi.org/10.48550/arXiv.2303.15770
    [109] W. Wei, E. Poirion, B. Bodini, M. Tonietto, S. Durrleman, O. Colliot, et al., Predicting PET-derived myelin content from multisequence MRI for individual longitudinal analysis in multiple sclerosis, Neuroimage, 223 (2020), 117308. https://doi.org/10.1016/j.neuroimage.2020.117308 doi: 10.1016/j.neuroimage.2020.117308
    [110] Q. Hu, H. Li, J. Zhang, Domain-adaptive 3D medical image synthesis: An efficient unsupervised approach, in International Conference on Medical Image Computing and Computer-Assisted Intervention, (2022), 495–504. https://doi.org/10.1007/978-3-031-16446-0_47
    [111] X. Meng, Y. Gu, Y. Pan, N. Wang, P. Xue, M. Lu, et al., A novel unified conditional score-based generative framework for multi-modal medical image completion, preprint, arXiv: 2207.03430. https://doi.org/10.48550/arXiv.2207.03430
    [112] V. Bharti, B. Biswas, K. K. Shukla, Qemcgan: quantized evolutionary gradient aware multiobjective cyclic gan for medical image translation, IEEE J. Biomed. Health Inf., 2023. https://doi.org/10.1109/JBHI.2023.3263434 doi: 10.1109/JBHI.2023.3263434
    [113] O. S. Al-Kadi, I. Almallahi, A. Abu-Srhan, A. M. Abushariah, W. Mahafza, Unpaired MR-CT brain dataset for unsupervised image translation, Data Brief, 42 (2022), 108109. https://doi.org/10.1016/j.dib.2022.108109 doi: 10.1016/j.dib.2022.108109
    [114] B. H. Menze, A. Jakab, S. Bauer, J. Kalpathy-Cramer, K. Farahani, J. Kirby, et al., The multimodal brain tumor image segmentation benchmark (BRATS), IEEE Trans. Med. Imaging, 34 (2014), 1993–2024. https://doi.org/10.1109/TMI.2014.2377694 doi: 10.1109/TMI.2014.2377694
    [115] T. Nyholm, S. Svensson, S. Andersson, J. Jonsson, M. Sohlin, C. Gustafsson, et al., MR and CT data with multiobserver delineations of organs in the pelvic area—Part of the Gold Atlas project, Med. Phys., 45 (2018), 1295–1300. https://doi.org/10.1002/mp.12748 doi: 10.1002/mp.12748
    [116] L. Jiang, Y. Mao, X. Chen, X. Wang, C. Li, CoLa-Diff: Conditional latent diffusion model for multi-modal MRI synthesis, preprint, arXiv: 2303.14081. https://doi.org/10.48550/arXiv.2303.14081
    [117] M. Özbey, O. Dalmaz, S. U. Dar, H. A. Bedel, Ş. Özturk, A. Güngör, et al., Unsupervised medical image translation with adversarial diffusion models, IEEE Trans. Med. Imaging, 2023. https://doi.org/10.1109/TMI.2023.3290149 doi: 10.1109/TMI.2023.3290149
    [118] J. Peng, R. L. Qiu, J. F. Wynne, C. Chang, S. Pan, T. Wang, et al., CBCT-based synthetic CT image generation using conditional denoising diffusion probabilistic model, preprint, arXiv: 2303.02649. https://doi.org/10.48550/arXiv.2303.02649
    [119] Q. Lyu, G. Wang, Conversion between CT and MRI images using diffusion and score-matching models, preprint, arXiv: 2209.12104. https://doi.org/10.48550/arXiv.2209.12104
    [120] S. Pan, E. Abouei, J. Wynne, T. Wang, R. L. Qiu, Y. Li, et al., Synthetic CT generation from MRI using 3D transformer-based denoising diffusion model, preprint, arXiv: 2305.19467. https://doi.org/10.48550/arXiv.2305.19467
    [121] F. Bazangani, F. J. Richard, B. Ghattas, E. Guedj, FDG-PET to T1 weighted MRI translation with 3D elicit generative adversarial network (E-GAN), Sensors, 22 (2022), 4640. https://doi.org/10.3390/s22124640 doi: 10.3390/s22124640
    [122] H. Emami, M. Dong, C. Glide-Hurst, CL-GAN: Contrastive learning-based generative adversarial network for modality transfer with limited paired data, in European Conference on Computer Vision, (2022), 527–542. https://doi.org/10.1007/978-3-031-25066-8_30
    [123] I. S. A. Abdelhalim, M. F. Mohamed, Y. B. Mahdy, Data augmentation for skin lesion using self-attention based progressive generative adversarial network, Expert Syst. Appl., 165 (2021), 113922. https://doi.org/10.1016/j.eswa.2020.113922 doi: 10.1016/j.eswa.2020.113922
    [124] A. A. E. Ambita, E. N. V. Boquio, P. C. Naval Jr, Covit-gan: vision transformer forcovid-19 detection in CT scan imageswith self-attention GAN for data augmentation, in International Conference on Artificial Neural Networks, (2021), 587–598. https://doi.org/10.1007/978-3-030-86340-1_47
    [125] M. Hajij, G. Zamzmi, R. Paul, L. Thukar, Normalizing flow for synthetic medical images generation, in 2022 IEEE Healthcare Innovations and Point of Care Technologies (HI-POCT), (2022), 46–49. https://doi.org/10.1109/HI-POCT54491.2022.9744072
    [126] R. Summers, Nih chest x-ray dataset of 14 common thorax disease categories, NIH Clinical Center: Bethesda, MD, USA, 2019.
    [127] P. A. Moghadam, S. V. Dalen, K. C. Martin, J. Lennerz, S. Yip, H. Farahani, et al., A morphology focused diffusion probabilistic model for synthesis of histopathology images, in Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, (2023), 2000–2009. https://doi.org/10.1109/WACV56688.2023.00204
    [128] S. Shahriar, S. Allana, M. H. Fard, R. Dara. A survey of privacy risks and mitigation strategies in the artificial intelligence life cycle, IEEE Access, 2023. https://doi.org/10.1109/ACCESS.2023.3287195 doi: 10.1109/ACCESS.2023.3287195
    [129] R. L. Grossman, A. P. Heath, V. Ferretti, H. E. Varmus, D. R. Lowy, W. A. Kibbe, et al., Toward a shared vision for cancer genomic data, N. Engl. J. Med., 375 (2016), 1109–1112. https://doi.org/10.1056/NEJMp1607591 doi: 10.1056/NEJMp1607591
    [130] S. Pan, T. Wang, R. L. Qiu, M. Axente, C. Chang, J. Peng, et al., 2D medical image synthesis using transformer-based denoising diffusion probabilistic model, Phys. Med. Biol., 68 (2023), 105004. https://doi.org/10.1088/1361-6560/acca5c doi: 10.1088/1361-6560/acca5c
    [131] X. Wang, Y. Peng, L. Lu, Z. Lu, M. Bagheri, R. M. Summers, Chestx-ray8: Hospital-scale chest x-ray database and benchmarks on weakly-supervised classification and localization of common thorax diseases, in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, (2017), 2097–2106. https://doi.org/10.1109/CVPR.2017.369
    [132] O. Bernard, A. Lalande, C. Zotti, F. Cervenansky, X. Yang, P. Heng, et al., Deep learning techniques for automatic MRI cardiac multi-structures segmentation and diagnosis: is the problem solved?, IEEE Trans. Med. Imaging, 37 (2018), 2514–2525. https://doi.org/10.1109/TMI.2018.2837502 doi: 10.1109/TMI.2018.2837502
    [133] B. Landman, Z. Xu, J. E. Igelsias, M. Styner, T. R. Langerak, A. Klein, 2015 miccai multi-atlas labeling beyond the cranial vault workshop and challenge, in Proc. MICCAI Multi-Atlas Labeling Beyond Cranial Vault—Workshop Challenge, 2015.
    [134] R. Zhang, W. Lu, J. Gao, Y. Tian, X. Wei, C. Wang, et al., RFI-GAN: A reference-guided fuzzy integral network for ultrasound image augmentation, Inf. Sci., 623 (2023), 709–728. https://doi.org/10.1016/j.ins.2022.12.026 doi: 10.1016/j.ins.2022.12.026
    [135] R. Zhang, W. Lu, X. Wei, J. Zhu, H. Jiang, Z. Liu, et al., A progressive generative adversarial method for structurally inadequate medical image data augmentation, IEEE J. Biomed. Health Inf., 26 (2021), 7–16. https://doi.org/10.1109/JBHI.2021.3101551 doi: 10.1109/JBHI.2021.3101551
    [136] K. Guo, J. Chen, T. Qiu, S. Guo, T. Luo, T. Chen, et al., MedGAN: An adaptive GAN approach for medical image generation, Comput. Biol. Med., (2023), 107119. https://doi.org/10.1016/j.compbiomed.2023.107119 doi: 10.1016/j.compbiomed.2023.107119
    [137] B. Kim, J. C. Ye, Diffusion deformable model for 4D temporal medical image generation, in International Conference on Medical Image Computing and Computer-Assisted Intervention, (2022), 539–548. https://doi.org/10.1007/978-3-031-16431-6_51
    [138] W. H. Pinaya, P. Tudosiu, J. Dafflon, P. F. D. Costa, V. Fernandez, P. Nachev, et al., Brain imaging generation with latent diffusion models, in MICCAI Workshop on Deep Generative Models, (2022), 117–126. https://doi.org/10.1007/978-3-031-18576-2_12
    [139] P. Tschandl, C. Rosendahl, H. Kittler, The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions, Sci. Data, 5 (2018), 1–9. https://doi.org/10.1038/sdata.2018.161 doi: 10.1038/sdata.2018.161
    [140] J. Nada, S. Bougleux, J. Lapuyade-Lahorgue, S. Ruan, F. Ghazouani, MR image synthesis using Riemannian geometry constrained in VAE, in 2022 16th IEEE International Conference on Signal Processing (ICSP), (2022), 485–488. https://doi.org/10.1109/ICSP56322.2022.9965357
    [141] H. Dai, Z. Liu, W. Liao, X. Huang, Y. Cao, Z. Wu, et al., AugGPT: Leveraging ChatGPT for text data augmentation, preprint, arXiv: 2302.13007. https://doi.org/10.48550/arXiv.2302.13007
    [142] H. Li, Y. Wu, V. Schlegel, R. Batista-Navarro, T. Nguyen, A. R. Kashyap, et al., PULSAR: Pre-training with extracted healthcare terms for summarising patients' problems and data augmentation with black-box large language models, preprint, arXiv: 2306.02754. https://doi.org/10.48550/arXiv.2306.02754
    [143] D. Jin, E. Pan, N. Oufattole, W. Weng, H. Fang, P. Szolovits, What disease does this patient have? a large-scale open domain question answering dataset from medical exams, Appl. Sci., 11 (2021), 6421. https://doi.org/10.3390/app11146421 doi: 10.3390/app11146421
    [144] A. Pal, L. K. Umapathi, M. Sankarasubbu, Medmcqa: A large-scale multi-subject multi-choice dataset for medical domain question answering, in Conference on Health, Inference, and Learning, (2022), 248–260.
    [145] D. Hendrycks, C. Burns, S. Basart, A. Zou, M. Mazeika, D. Song, et al. Measuring massive multitask language understanding, preprint, arXiv: 2009.03300. https://doi.org/10.48550/arXiv.2009.03300
    [146] Q. Jin, B. Dhingra, Z. Liu, W. W. Cohen, X. Lu, Pubmedqa: A dataset for biomedical research question answering, preprint, arXiv: 1909.06146. https://doi.org/10.48550/arXiv.1909.06146
    [147] A. B. Abacha, E. Agichtein, Y. Pinter, D. Demner-Fushman, Overview of the medical question answering task at TREC 2017 LiveQA, in TREC, (2017), 1–12.
    [148] A. B. Abacha, Y. Mrabet, M. Sharp, T. R. Goodwin, S. E. Shooshan, D. Demner-Fushman, Bridging the gap between consumers' medication questions and trusted answers., in MedInfo, (2019), 25–29.
    [149] K. Singhal, S. Azizi, T. Tu, S. S. Mahdavi, J. Wei, H. W. Chung, et al., Large language models encode clinical knowledge, preprint, arXiv: 2212.13138. https://doi.org/10.48550/arXiv.2212.13138
    [150] A. Chowdhery, S. Narang, J. Devlin, M. Bosma, G. Mishra, A. Roberts, et al., Palm: Scaling language modeling with pathways, preprint, arXiv: 2204.02311. https://doi.org/10.48550/arXiv.2204.02311
    [151] C. Wu, X. Zhang, Y. Zhang, Y. Wang, W. Xie, Pmc-llama: Further finetuning llama on medical papers, preprint, arXiv: 2304.14454. https://doi.org/10.48550/arXiv.2304.14454
    [152] H. Touvron, T. Lavril, G. Izacard, X. Martinet, M. Lachaux, T. Lacroix, et al., Llama: Open and efficient foundation language models, preprint, arXiv: 2302.13971. https://doi.org/10.48550/arXiv.2302.13971
    [153] K. Lo, L. L. Wang, M. Neumann, R. Kinney, D. S. Weld, S2ORC: The semantic scholar open research corpus, preprint, arXiv: 1911.02782. https://doi.org/10.48550/arXiv.1911.02782
    [154] O. Thawkar, A. Shaker, S. S. Mullappilly, H. Cholakkal, R. M. Anwer, S. Khan, et al., Xraygpt: Chest radiographs summarization using medical vision-language models, preprint, arXiv: 2306.07971. https://doi.org/10.48550/arXiv.2306.07971
    [155] W. Chiang, Z. Li, Z. Lin, Y. Sheng, Z. Wu, H. Zhang, et al., Vicuna: An open-source chatbot impressing gpt-4 with 90%* chatgpt quality, Available from: https://vicuna.lmsys.org.
    [156] A. E. Johnson, T. J. Pollard, S. J. Berkowitz, N. R. Greenbaum, M. P. Lungren, C. Deng, et al., MIMIC-CXR, a de-identified publicly available database of chest radiographs with free-text reports, Sci. Data, 6 (2019), 317. https://doi.org/10.1038/s41597-019-0322-0 doi: 10.1038/s41597-019-0322-0
    [157] D. Demner-Fushman, M. D. Kohli, M. B. Rosenman, S. E. Shooshan, L. Rodriguez, S. Antani, et al., Preparing a collection of radiology examinations for distribution and retrieval, J. Am. Med. Inf. Assoc., 23 (2016), 304–310. https://doi.org/10.1093/jamia/ocv080 doi: 10.1093/jamia/ocv080
    [158] J. Zhou, X. He, L. Sun, J. Xu, X. Chen, Y. Chu, et al., SkinGPT-4: An interactive dermatology diagnostic system with visual large language model, medRxiv, (2023), 2023–2026.
    [159] R. Daneshjou, M. Yuksekgonul, Z. R. Cai, R. Novoa, J. Y. Zou, Skincon: A skin disease dataset densely annotated by domain experts for fine-grained debugging and analysis, Adv. Neural Inf. Process. Syst., 35 (2022), 18157–18167.
    [160] D. Zhu, J. Chen, X. Shen, X. Li, M. Elhoseiny, Minigpt-4: Enhancing vision-language understanding with advanced large language models, preprint, arXiv: 2304.10592. https://doi.org/10.48550/arXiv.2304.10592
    [161] G. Zeng, W. Yang, Z. Ju, Y. Yang, S. Wang, R. Zhang, et al., MedDialog: Large-scale medical dialogue datasets, in Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), (2020), 9241–9250. https://doi.org/10.18653/v1/2020.emnlp-main.743
    [162] A. B. Abacha, Y. Mrabet, M. Sharp, T. R. Goodwin, S. E. Shooshan, D. Demner-Fushman, Bridging the Gap Between Consumers' Medication Questions and Trusted Answers, in MedInfo, (2019), 25–29.
    [163] M. Savery, A. B. Abacha, S. Gayen, D. Demner-Fushman, Question-driven summarization of answers to consumer health questions, Sci. Data, 7 (2020), 322. https://doi.org/10.1038/s41597-020-00667-z doi: 10.1038/s41597-020-00667-z
    [164] H. Yuan, Z. Yuan, R. Gan, J. Zhang, Y. Xie, S. Yu, BioBART: Pretraining and evaluation of a biomedical generative language model, preprint, arXiv: 2204.03905. https://doi.org/10.48550/arXiv.2204.03905
    [165] R. Luo, L. Sun, Y. Xia, T. Qin, S. Zhang, H. Poon, et al., BioGPT: generative pre-trained transformer for biomedical text generation and mining, Brief. BioInf., 23 (2022), bbac409. https://doi.org/10.1093/bib/bbac409 doi: 10.1093/bib/bbac409
    [166] J. Li, Y. Sun, R. J. Johnson, D. Sciaky, C. Wei, R. Leaman, et al., BioCreative V CDR task corpus: a resource for chemical disease relation extraction, Database, 2016 (2016). https://doi.org/10.1093/database/baw068 doi: 10.1093/database/baw068
    [167] Y. Hou, Y. Xia, L. Wu, S. Xie, Y. Fan, J. Zhu, et al., Discovering drug-target interaction knowledge from biomedical literature, Bioinformatics, 38 (2022), 5100–5107. https://doi.org/10.1093/bioinformatics/btac648 doi: 10.1093/bioinformatics/btac648
    [168] M. Herrero-Zazo, I. Segura-Bedmar, P. Martínez, T. Declerck, The DDI corpus: An annotated corpus with pharmacological substances and drug–drug interactions, J. Biomed. Inf., 46 (2013), 914–920. https://doi.org/10.1016/j.jbi.2013.07.011 doi: 10.1016/j.jbi.2013.07.011
    [169] S. Baker, I. Silins, Y. Guo, I. Ali, J. Högberg, U. Stenius, et al., Automatic semantic classification of scientific literature according to the hallmarks of cancer, Bioinformatics, 32 (2016), 432–440. https://doi.org/10.1093/bioinformatics/btv585 doi: 10.1093/bioinformatics/btv585
    [170] A. Venigalla, J. Frankle, M. Carbin, Biomedlm: a domain-specific large language model for biomedical text, MosaicML. Accessed: Dec, 23 (2022), 2.
    [171] G. Balikas, A. Krithara, I. Partalas, G. Paliouras, Bioasq: A challenge on large-scale biomedical semantic indexing and question answering, in Multimodal Retrieval in the Medical Domain: First International Workshop, MRMD 2015, Vienna, Austria, March 29, 2015, Revised Selected Papers, (2015), 26–39. https://doi.org/10.1007/978-3-319-24471-6_3
    [172] A. B. Abacha, Y. M Rabet, Y. Zhang, C. Shivade, C. Langlotz, D. Demner-Fushman, Overview of the MEDIQA 2021 shared task on summarization in the medical domain, in Proceedings of the 20th Workshop on Biomedical Language Processing, (2021), 74–85. https://doi.org/10.18653/v1/2021.bionlp-1.8
    [173] S. Mohan, D. Li, Medmentions: A large biomedical corpus annotated with umls concepts, preprint, arXiv: 1902.09476. https://doi.org/10.48550/arXiv.1902.09476
    [174] R. I. Doğan, R. Leaman, Z. Lu, NCBI disease corpus: a resource for disease name recognition and concept normalization, J. Biomed. Inf., 47 (2014), 1–10. https://doi.org/10.1016/j.jbi.2013.12.006 doi: 10.1016/j.jbi.2013.12.006
    [175] M. Basaldella, F. Liu, E. Shareghi, N. Collier, COMETA: A corpus for medical entity linking in the social media, preprint, arXiv: 2010.03295. https://doi.org/10.48550/arXiv.2010.03295
    [176] N. Limsopatham, N. Collier, Normalising medical concepts in social media texts by learning semantic representation, in Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long papers), (2016), 1014–1023. https://doi.org/10.18653/v1/P16-1096
    [177] S. Pradhan, N. Elhadad, B. R. South, D. Martinez, L. M. Christensen, A. Vogel, et al., Task 1: ShARe/CLEF eHealth Evaluation Lab 2013., CLEF (working notes), 1179 (2013).
    [178] D. L. Mowery, S. Velupillai, B. R. South, L. Christensen, D. Martinez, L. Kelly, et al., Task 2: ShARe/CLEF eHealth evaluation lab 2014, in Proceedings of CLEF 2014, (2014).
    [179] S. Karimi, A. Metke-Jimenez, M. Kemp, C. Wang, Cadec: A corpus of adverse drug event annotations, J. Biomed. Inf., 55 (2015), 73–81. https://doi.org/10.1016/j.jbi.2015.03.010 doi: 10.1016/j.jbi.2015.03.010
    [180] J. Kim, T. Ohta, Y. Tateisi, J. I. Tsujii, GENIA corpus—a semantically annotated corpus for bio-textmining, Bioinformatics, 19 (2003), i180–i182. https://doi.org/10.1093/bioinformatics/btg1023 doi: 10.1093/bioinformatics/btg1023
    [181] Y. Li, Z. Li, K. Zhang, R. Dan, Y. Zhang, Chatdoctor: A medical chat model fine-tuned on llama model using medical domain knowledge, preprint, arXiv: 2303.14070. https://doi.org/10.48550/arXiv.2303.14070
    [182] A. Toma, P. R. Lawler, J. Ba, R. G. Krishnan, B. B. Rubin, B. Wang, Clinical camel: An open-source expert-level medical language model with dialogue-based knowledge encoding, preprint, arXiv: 2305.12031. https://doi.org/10.48550/arXiv.2305.12031
    [183] G. Wang, G. Yang, Z. Du, L. Fan, X. Li, ClinicalGPT: Large language models finetuned with diverse medical data and comprehensive evaluation, preprint, arXiv: 2306.09968. https://doi.org/10.48550/arXiv.2306.09968
    [184] S. Zhang, X. Zhang, H. Wang, L. Guo, S. Liu, Multi-scale attentive interaction networks for chinese medical question answer selection, IEEE Access, 6 (2018), 74061–74071. https://doi.org/10.1109/ACCESS.2018.2883637 doi: 10.1109/ACCESS.2018.2883637
    [185] T. M. Lai, C. Zhai, H. Ji, KEBLM: Knowledge-enhanced biomedical language models, J. Biomed. Inf., 143 (2023), 104392. https://doi.org/10.1016/j.jbi.2023.104392 doi: 10.1016/j.jbi.2023.104392
    [186] J. Lee, W. Yoon, S. Kim, D. Kim, S. Kim, C. H. So, et al., BioBERT: a pre-trained biomedical language representation model for biomedical text mining, Bioinformatics, 36 (2020), 1234–1240. https://doi.org/10.1093/bioinformatics/btz682 doi: 10.1093/bioinformatics/btz682
    [187] I. Beltagy, K. Lo, A. Cohan, SciBERT: A pretrained language model for scientific text, preprint, arXiv: 1903.10676. https://doi.org/10.48550/arXiv.1903.10676
    [188] A. Romanov, C. Shivade, Lessons from natural language inference in the clinical domain, preprint, arXiv: 1808.06752. https://doi.org/10.48550/arXiv.1808.06752
    [189] H. W. Chung, L. Hou, S. Longpre, B. Zoph, Y. Tay, W. Fedus, et al., Scaling instruction-finetuned language models, preprint, arXiv: 2210.11416. https://doi.org/10.48550/arXiv.2210.11416
    [190] Y. Gao, T. Miller, M. Afshar, D. Dligach, BioNLP Workshop 2023 Shared Task 1A: Problem List Summarization, in Proceedings of the 22nd Workshop on Biomedical Language Processing, 2023.
    [191] J. Hu, Z. Li, Z. Chen, Z. Li, X. Wan, T. Chang, Graph enhanced contrastive learning for radiology findings summarization, preprint, arXiv: 2204.00203. https://doi.org/10.48550/arXiv.2204.00203
    [192] C. Ma, Z. Wu, J. Wang, S. Xu, Y. Wei, Z. Liu, et al., ImpressionGPT: an iterative optimizing framework for radiology report summarization with chatGPT, preprint, arXiv: 2304.08448. https://doi.org/10.48550/arXiv.2304.08448
    [193] B. Pang, E. Nijkamp, W. Kryściński, S. Savarese, Y. Zhou, C. Xiong, Long document summarization with top-down and bottom-up inference, preprint, arXiv: 2203.07586. https://doi.org/10.48550/arXiv.2203.07586
    [194] G. Frisoni, P. Italiani, S. Salvatori, G. Moro, Cogito ergo summ: abstractive summarization of biomedical papers via semantic parsing graphs and consistency rewards, in Proceedings of the AAAI Conference on Artificial Intelligence, (2023), 12781–12789. https://doi.org/10.1609/aaai.v37i11.26503
    [195] Y. Guo, W. Qiu, Y. Wang, T. Cohen, Automated lay language summarization of biomedical scientific reviews, in Proceedings of the AAAI Conference on Artificial Intelligence, (2021), 160–168. https://doi.org/10.1609/aaai.v35i1.16089
    [196] S. Casper, X. Davies, C. Shi, T. K. Gilbert, J. Scheurer, J. Rando, et al., Open problems and fundamental limitations of reinforcement learning from human feedback, preprint, arXiv: 2307.15217. https://doi.org/10.48550/arXiv.2307.15217
    [197] O. Ostapenko, T. Lesort, P. Rodriguez, M. R. Arefin, A. Douillard, I. Rish, et al., Continual learning with foundation models: An empirical study of latent replay, in Conference on Lifelong Learning Agents, (2022), 60–91.
    [198] I. Chalkidis, M. Fergadiotis, P. Malakasiotis, N. Aletras, I. Androutsopoulos, LEGAL-BERT: The muppets straight out of law school, preprint, arXiv: 2010.02559. https://doi.org/10.48550/arXiv.2010.02559
    [199] J. Hoffmann, S. Borgeaud, A. Mensch, E. Buchatskaya, T. Cai, E. Rutherford, et al., Training compute-optimal large language models, preprint, arXiv: 2203.15556. https://doi.org/10.48550/arXiv.2203.15556
    [200] A. Aghajanyan, L. Yu, A. Conneau, W. Hsu, K. Hambardzumyan, S. Zhang, et al., Scaling laws for generative mixed-modal language models, preprint, arXiv: 2301.03728. https://doi.org/10.48550/arXiv.2301.03728
    [201] D. Shah, H. A. Schwartz, D. Hovy, Predictive biases in natural language processing models: A conceptual framework and overview, preprint, arXiv: 2301.03728. https://doi.org/10.48550/arXiv.2301.03728
    [202] Y. Dong, N. Liu, B. Jalaian, J. Li, Edits: Modeling and mitigating data bias for graph neural networks, in Proceedings of the ACM Web Conference 2022, (2022), 1259–1269. https://doi.org/10.1145/3485447.3512173
    [203] H. Zhao, W. Zhou, D. Chen, T. Wei, W. Zhang, N. Yu, Multi-attentional deepfake detection, in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, (2021), 2185–2194. https://doi.org/10.1109/CVPR46437.2021.00222
    [204] A. Brauneck, L. Schmalhorst, M. M. K. Majdabadi, M. Bakhtiari, U. Völker, J. Baumbach, et al., Federated machine learning, privacy-enhancing technologies, and data protection laws in medical research: Scoping review, J. Med. Internet Res., 25 (2023), e41588. https://doi.org/10.2196/41588 doi: 10.2196/41588
    [205] Q. Yang, Y. Liu, T. Chen, Y. Tong, Federated machine learning: Concept and applications, ACM Trans. Intell. Syst. Technol., 10 (2019), 1–19. https://doi.org/10.1145/3298981 doi: 10.1145/3298981
    [206] P. Zhang, M. N. K. Boulos, Generative AI in medicine and healthcare: promises, opportunities and challenges, Future Internet, 15 (2023), 286. https://doi.org/10.3390/fi15090286 doi: 10.3390/fi15090286
  • Reader Comments
  • © 2024 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Metrics

Article views(6670) PDF downloads(1198) Cited by(12)

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog