Is ChatGPT able to interpret arterial blood gas analysis?A comparative cross-sectional study

Background: This study aimed to evaluate the performance of ChatGPT in interpreting arterial blood gas results and compare it to the performance of physicians with varying levels of experience. Methods: In this comparative and cross-sectional study, 30 selected clinical cases encompassing simple and mixed acid-base disorders, respiratory abnormalities, and electrolyte disturbances, were analyzed by ChatGPT and 45 anesthesiology physicians who were divided into three groups: specialists, experienced residents, and inexperienced residents. Participants assessed arterial blood gases across five domains, including acid-base diagnosis, respiratory evaluation, fluid and electrolyte disturbance, other abnormalities, and treatment planning. Both ChatGPT and participant responses were scored by two experienced anesthesiology academicians based on a 5-point Likert scale. The overall score was calculated as the sum of the scores from the five individual subdomains. Results: The overall scores of ChatGPT and the physicians were similar (21.75, 21.97). ChatGPT demonstrated high accuracy in diagnosing primary acid-base disorders and providing treatment recommendations, though slight variability was observed in mixed disorders. A strong correlation was observed between ChatGPT’s scores and those of physicians (r = 0.912, p < 0.001). Conclusions: ChatGPT demonstrated a performance comparable to that of physicians, suggesting that it could assist in the decision-making processes of healthcare professionals and contribute to workload reduction.

Artificial intelligence; Arterial blood gas; Critical care; Anesthesia; ChatGPT; Acid-base disorders; Decision support

[1] Weimar Z, Smallwood N, Shao J, Chen XE, Moran TP, Khor YH. Arterial blood gas analysis or venous blood gas analysis for adult hospitalised patients with respiratory presentations: a systematic review. Internal Medicine Journal. 2024; 54: 1531–1540.

[2] Korpi-Steiner N, Horowitz G, Tesfazghi M, Suh-Lailam BB. Current issues in blood gas analysis. The Journal of Applied Laboratory Medicine. 2023; 8: 372–381.

[3] Mohammed HM, Abdelatief DA. Easy blood gas analysis: implications for nursing. Egyptian Journal of Chest Diseases and Tuberculosis. 2016; 65: 369–376.

[4] Rodríguez-Villar S, Poza-Hernández P, Freigang S, Zubizarreta-Ormazabal I, Paz-Martín D, Holl E, et al. Automatic real-time analysis and interpretation of arterial blood gas sample for point-of-care testing: clinical validation. PLOS ONE. 2021; 16: e0248264.

[5] Musa Hussain EY, Sidahmed Abdullah AM, Mahgoub Idris RM, Hashim Gabir ZT, Mohammed Diab RA, Mustafa Ahmed RN, et al. Evaluation and improving the quality of arterial blood gas interpretation among junior doctors in Aswan University Hospital: a clinical audit. Cureus. 2024; 16: e74906.

[6] Rizwan A, Sadiq T. The use of AI in diagnosing diseases and providing management plans: a consultation on cardiovascular disorders with ChatGPT. Cureus. 2023; 15: e43106.

[7] Mu Y, He D. The potential applications and challenges of ChatGPT in the medical field. International Journal of General Medicine. 2024; 17: 817–826.

[8] Scherr R, Halaseh FF, Spina A, Andalib S, Rivera R. ChatGPT interactive medical simulations for early clinical education: case study. JMIR Medical Education. 2023; 9: e49877.

[9] Puleio F, Lo Giudice G, Bellocchio AM, Boschetti CE, Lo Giudice R. Clinical, research, and educational applications of ChatGPT in dentistry: a narrative review. Applied Sciences. 2024; 14: 10802.

[10] Najjar R. Redefining radiology: a review of artificial intelligence integration in medical imaging. Diagnostics. 2023; 13: 2760.

[11] Försch S, Klauschen F, Hufnagl P, Roth W. Artificial intelligence in pathology. Deutsches Ärzteblatt International. 2021; 118: 194–204.

[12] Irfan B, Yaqoob A. ChatGPT’s epoch in rheumatological diagnostics: a critical assessment in the context of Sjögren’s Syndrome. Cureus. 2023; 15: e47754.

[13] Köroğlu EY, Fakı S, Beştepe N, Tam AA, Çuhacı Seyrek N, Topaloglu O, et al. A novel approach: evaluating ChatGPT’s utility for the management of thyroid nodules. Cureus. 2023; 15: e47576.

[14] Oca MC, Meller L, Wilson K, Parikh AO, McCoy A, Chang J, et al. Bias and Inaccuracy in AI Chatbot ophthalmologist recommendations. Cureus. 2023; 15: e45911.

[15] Puladi B, Gsaxner C, Kleesiek J, Hölzle F, Röhrig R, Egger J. The impact and opportunities of large language models like ChatGPT in oral and maxillofacial surgery: a narrative review. International Journal of Oral and Maxillofacial surgery. 2024; 53: 78–88.

[16] Sallam M, Al-Salahat K. Below average ChatGPT performance in medical microbiology exam compared to university students. Frontiers in Education. 2023; 8: 1333415.

[17] Hadi A, Tran E, Nagarajan B, Kirpalani A. Evaluation of ChatGPT as a diagnostic tool for medical learners and clinicians. PLOS ONE. 2024; 19: e0307383.

[18] Scheschenja M, Viniol S, Bastian MB, Wessendorf J, König AM, Mahnken AH. Feasibility of GPT-3 and GPT-4 for in-depth patient education prior to interventional radiological procedures: a comparative analysis. Cardiovascular and Interventional Radiology. 2024; 47: 245–250.

[19] Tangadulrat P, Sono S, Tangtrakulwanich B. Using ChatGPT for clinical practice and medical education: cross-sectional survey of medical students’ and physicians’ perceptions. JMIR Medical Education. 2023; 9: e50658.

[20] Cocci A, Pezzoli M, Lo Re M, Russo GI, Asmundo MG, Fode M, et al. Quality of information and appropriateness of ChatGPT outputs for urology patients. Prostate Cancer and Prostatic Diseases. 2024; 27: 103–108.

[21] Marino PL. Chapter 33: Acid-Base Disorders. Marino’s The ICU Book (pp. 544–557). 5th edn. Philadelphia: Lippincott Williams & Wilkins; 2024.

[22] Kidney Disease: Improving Global Outcomes (KDIGO) CKD Work Group. KDIGO 2024 clinical practice guideline for the evaluation and management of chronic kidney disease. Kidney International. 2024; 105: S117–S314.

[23] Cederholm T, Barazzoni R, Austin P, Gonzalez MC, Fukushima R, Correia MITD, et al. GLIM criteria for the diagnosis of malnutrition – A consensus report from the global clinical nutrition community*. Journal of Cachexia, Sarcopenia and Muscle. 2019; 10: 207–217.

[24] McDonagh TA, Metra M, Adamo M, Gardner RS, Baumbach A, Böhm M, et al.; ESC Scientific Document Group. 2021 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure. European Heart Journal. 2021; 42: 3599–3726.

[25] ElSayed NA, Aleppo G, Aroda VR, Bannuru RR, Brown FM, Bruemmer D, et al. 6. Glycemic targets: standards of care in diabetes—2023. Diabetes Care. 2023; 46: S97–S110.

[26] Sood P, Paul G, Puri S. Interpretation of arterial blood gas. Indian Journal of Critical Care Medicine. 2010; 14: 57–64.

[27] Evans L, Rhodes A, Alhazzani W, Antonelli M, Coopersmith CM, French C, et al. Surviving sepsis campaign: international guidelines for management of sepsis and septic shock 2021. Intensive Care Medicine. 2021; 47: 1181–1247.

[28] Iolascon A, Bianchi P, Andolfo I, Russo R, Barcellini W, Fermo E, et al.; SWG of red cell and iron of EHA and EuroBloodNet. Recommendations for diagnosis and treatment of methemoglobinemia. American Journal of Hematology. 2021; 96: 1666–1678.

[29] Yee J, Frinak S, Mohiuddin N, Uduman J. Fundamentals of arterial blood gas interpretation. Kidney360. 2022; 3: 1458–1466.

[30] Cross JL, Choma MA, Onofrey JA. Bias in medical AI: implications for clinical decision-making. PLOS Digital Health. 2024; 3: e0000651.

[31] Wong A, Otles E, Donnelly JP, Krumm A, McCullough J, DeTroyer-Cooley O, et al. External validation of a widely implemented proprietary sepsis prediction model in hospitalized patients. JAMA Internal Medicine. 2021; 181: 1065–1070.

[32] Hirosawa T, Suzuki T, Shiraishi T, Hayashi A, Fujii Y, Harada T, et al. Adapting artificial intelligence concepts to enhance clinical decision-making: a hybrid intelligence framework. International Journal of General Medicine. 2024; 17: 5417–5422.

[33] Patel BN, Rosenberg L, Willcox G, Baltaxe D, Lyons M, Irvin J, et al. Erratum: author correction: human-machine partnership with artificial intelligence for chest radiograph diagnosis. NPJ Digital Medicine. 2019; 2: 129.

[34] Wang S, Zhao Z, Ouyang X, Liu T, Wang Q, Shen D. Interactive computer-aided diagnosis on medical image using large language models. Communications Engineering. 2024; 3: 133.

[35] Lee TC, Staller K, Botoman V, Pathipati MP, Varma S, Kuo B. ChatGPT answers common patient questions about colonoscopy. Gastroenterology. 2023; 165: 509–511.e7.

[36] Xue VW, Lei P, Cho WC. The potential impact of ChatGPT in clinical and translational medicine. Clinical and Translational Medicine. 2023; 13: e1216.

[37] Wang L, Zhang Z, Wang D, Cao W, Zhou X, Zhang P, et al. Human-centered design and evaluation of AI-empowered clinical decision support systems: a systematic review. Frontiers in Computer Science. 2023; 5: 1187299.

[38] Coskun AB, Elmaoglu E, Buran C, Alsaç SY. Integration of ChatGPT and e-health literacy: opportunities, challenges, and a look towards the future. Journal of Health Reports and Technology. 2024; 10: e139748.

[39] Koubaa A, Boulila W, Ghouti L, Alzahem A, Latif S. Exploring ChatGPT capabilities and limitations: a survey. IEEE Access. 2023; 11: 118698–118721.

[40] Sallam M, Salim NA, Barakat M, Al-Tammemi AB. ChatGPT applications in medical, dental, pharmacy, and public health education: a descriptive study highlighting the advantages and limitations. Narra J. 2023; 3: e103.

[41] Jones C, Thornton J, Wyatt JC. Artificial intelligence and clinical decision support: clinicians’ perspectives on trust, trustworthiness, and liability. Medical Law Review. 2023; 31: 501–520.

[42] Sallam M. The utility of ChatGPT as an example of large language models in healthcare education, research and practice: systematic review on the future perspectives and potential limitations. Healthcare. 2023; 11: 887.