A big congratulation is given to Dr. Li-Chun Liu as the 2021 Journal of the Chinese Medical Association Outstanding Research Paper Award winner who is selected from all authors having contributed their research works to the Journal of the Chinese Medical Association (JCMA) last year.1–6 This year's award, sponsored by the Chinese Medical Association-Taipei (CMA-Taipei), is granted to researchers who demonstrate their excellent performance and great contribution to enhanced better patients’ care. Dr. Liu won this credit at the Annual Meeting of the CMA-Taipei on July 16, 2022, held at Taipei, Taiwan, and the meeting is still affected apparently by the emergence of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) (coronavirus disease 2019 [COVID-19]) pandemic.7–9 Therefore, the annual conference is still held in the hybrid form, including virtual and physical meetings.

Dr. Liu and her colleagues attempted to use a novel artificial intelligence (AI) tool, such as fully convolutional networks (FCNs) and deep learning to resolve the limits of physicians who are unable to monitor immediate and complicated physical changes of human beings at every moment. The intrapartum cardiotocography, one of the best examples and also a target of Dr. Liu’s work, is successfully applied to simultaneously monitor fetal well-being (fetal heart rate) and adequate uterine contraction during labor in a long-term period.6,9 With the assistance of intrapartum cardiotocography, many ominous signs, such as uterine hyperstimulation and abnormal fetal heartbeats, could be detected if the appropriate interpretation is given.6,9,10 However, as shown by Dr. Liu, visual interpretation of cardiotocography may be delayed and has a discrepancy between observers, resulting in different decision-making.6 With the implication of AI-aided analysis of cardiotocography by standard criteria, the discrepancy may be minimized.6 In Dr. Liu’s study design, nonreassuring fetal status may be at higher risk of the development of large discrepancy between the different observers. Therefore, the authors evaluated the concordance of interpretation of cardiotocography between physicians and AI-based FCNs of deep learning. The results showed a significant match between physicians and AI model to identify the six fetal heart rate classes, suggesting that the AI-based FCNs of deep learning is feasible for monitoring intrapartum fetal heartbeats. Additionally, there is a greatest accuracy to recognize fetal heart rate acceleration on electronic fetal monitoring, although there is less accuracy to identify variable deceleration of fetal heart rate on electronic fetal monitoring. Finally, they found the sensitivity of nonreassuring fetal status evaluation by the AI model is better than with clinical practice, although the false-positive rate was higher in the AI model than in the clinical practice,6 suggesting that the AI model has a competitive ability to identify fetal heart rate patterns but to evaluate the nonreassuring fetal status is still unsatisfactory. However, their effort is worthy of big applause, since it is a milestone.

It is well known that the shortage of manpower and care providers in the rural area is still the biggest challenge in Taiwan.11 Therefore, it is an urgent need to provide more effective equipment to give more precise interpretation in taking care of pregnant women during labor. Furthermore, of the most critical point, a continuous work without interruption or any need for rest is also the strength of AI mode compared to physicians. It is worthy of our efforts to provide newborns or pregnant women away from dangers and to further decrease intrapartum-related morbidities and mortalities during labor. AI, although not new, may provide a better chance to achieve the aforementioned goal,12–15 based on their uncanny ability to learn about what’s in a signal or photograph, the meaning of spoken language and much more, just through exposure.12 However, there is no doubt that AI-assisting interpretation of intrapartum cardiotocography is still far from perfect and may lead to some errors, such as overdiagnosis resulting in overtreatment, which has been demonstrated in Dr. Liu’ study.6 Moreover, deep learning needs much data access and considers personal privacy. Sometimes, it is hard to convince the public to overcome this limitation; therefore, we don’t think that the AI-assistance system can totally replace traditional approaches or standard care yet.15 By contrast, AI-assistance systems may provide a rapid, prompt, immediate, uninterrupted, and possibly cost-effective supplement that may be most useful in crisis setting or in contexts where the traditional medical care system is out of date, or when there is shortage of medical care providers.11,15 All suggest that there is still a long way to go.

We congratulate Dr. Liu to win the 2021 Journal of the Chinese Medical Association Outstanding Research Paper Award again, and welcome more and more excellent articles to be published in the Journal of the Chinese Medical Association.16 According to Dr. Savage’s recommendation, understanding why a system made a certain diagnosis can help to convince all physicians that it is legitimate in a medical setting.12 Finally, we appreciate all support to the JCMA, which obtains a new impact factor of 3.396 in 2021. We encourage researchers to use this platform to publish their excellent works, since with inspiration and effort by readers, authors, reviewers, and editors, the evolution of medical sciences toward better care for patients can be easily achieved.

ACKNOWLEDGMENTS

This article was supported by grants from the Taiwan Ministry of Science and Technology, Executive Yuan, Taiwan (MOST 110-2314-B-075-016-MY3 and MOST 111-2314-B-075-045), and Taipei Veterans General Hospital (V110C-082, and VGH111C-103). The authors appreciate the support from Female Cancer Foundation, Taipei, Taiwan.

REFERENCES 1. Liu LC, Tsai YH, Chou YC, Jheng YC, Lin CK, Lyu NY, et al. Concordance analysis of intrapartum cardiotocography between physicians and artificial intelligence-based technique using modified one-dimensional fully convolutional networks. J Chin Med Assoc. 2021;84:158–64. 2. Huang HC, Hsu SJ, Chuang CL, Hsiung SY, Chang CC, Hou MC, et al. Effects of dipeptidyl peptidase-4 inhibition on portal hypertensive and cirrhotic rats. J Chin Med Assoc. 2021;84:1092–9. 3. Liu CH, Kung YH, Chien-Fu Lin J, Chuang CM, Wu HH, Jiang LY, et al. Synergistic therapeutic effect of low-dose bevacizumab with cisplatin-based chemotherapy for advanced or recurrent cervical cancer. J Chin Med Assoc. 2021;84:1139–44. 4. Lin WY, Chung FP, Liao CT, Huang JL, Liang HW, Lee YH, et al. Treatment with angiotensin receptor neprilysin inhibitor for Taiwan heart failure patients: rationale and baseline characteristics of the TAROT-HF study. J Chin Med Assoc. 2021;84:833–41. 5. Wang JW, Chen PY, Huang HH, Yeh C, Chen SC, Lee WJ, et al. Change of plasma amylin after bariatric surgery challenged by oral glucose is associated with remission of type 2 diabetes mellitus. J Chin Med Assoc. 2021;84:1001–6. 6. Huang LJ, Huang HC, Chuang CL, Chang SL, Tsai HC, Lu DY, et al. Role-play of real patients improves the clinical performance of medical students. J Chin Med Assoc. 2021;84:183–90. 7. Wang PH, Lee WL, Yang ST, Tsui KH, Chang CC, Lee FK. The impact of COVID-19 in pregnancy: part I. Clinical presentations and untoward outcomes of pregnant women with COVID-19. J Chin Med Assoc. 2021;84:813–20. 8. Yang ST, Yeh CC, Lee WL, Lee FK, Chang CC, Wang PH. A symptomatic near-term pregnant woman recovered from SARS-CoV-2 infection. Taiwan J Obstet Gynecol. 2021;60:945–8. 9. Wang PH, Lee WL, Yang ST, Tsui KH, Chang CC, Lee FK. The impact of COVID-19 in pregnancy: part II. Vaccination to pregnant women. J Chin Med Assoc. 2021;84:903–10. 10. Lee HH, Huang BS, Cheng M, Yeh CC, Lin IC, Horng HC, et al. Intracervical foley catheter plus intravaginal misoprostol vs intravaginal misoprostol alone for cervical ripening: a meta-analysis. Int J Environ Res Public Health. 2020;17:E1825. 11. Lee FK, Yang ST, Wang PH. No-fault compensation systems of childbirth accidents in Taiwan. Taiwan J Obstet Gynecol. 2022;61:409–10. 12. Savage N. Breaking into the black box of artificial intelligence [published online ahead of print March 29, 2022]. Nature. doi:10.1038/d41586-022-00858-1. 13. Chou TH, Yeh HJ, Chang CC, Tang JH, Kao WY, Su IC, et al. Deep learning for abdominal ultrasound: a computer-aided diagnostic system for the severity of fatty liver. J Chin Med Assoc. 2021;84:842–50. 14. Sun YC, Hsieh AT, Fang ST, Wu HM, Kao LW, Chung WY, et al. Can 3D artificial intelligence models outshine 2D ones in the detection of intracranial metastatic tumors on magnetic resonance images? J Chin Med Assoc. 2021;84:956–62. 15. Aiken E, Bellue S, Karlan D, Udry C, Blumenstock JE. Machine learning and phone data can improve targeting of humanitarian aid. Nature. 2022;603:864–70. 16. Wang PH, Huo TI. Outstanding research paper awards of the Journal of the Chinese Medical Association in 2020. J Chin Med Assoc. 2021;84:1071–2.