Artificial Intelligence in Intensive Care: Applications, Challenges, and Future Directions -A Review

Öz

Objective: Artificial intelligence (AI) has emerged as a transformative technology in intensive care units (ICUs), where clinicians must process large volumes of rapidly evolving physiological, laboratory, and imaging data. This review aims to evaluate current AI applications in critical care, highlight organ- and syndrome-specific use cases, identify major implementation challenges, and outline future directions necessary for safe and effective integration of AI into ICU practice. Method: A narrative review methodology was adopted. Relevant literature was identified through a non-systematic search of PubMed and major critical care journals, focusing on recent clinical, computational, and translational studies. Evidence was synthesized across functional domains—including diagnosis, risk stratification, prognostic modeling, decision support, and imaging analysis—and across organ-specific applications such as respiratory failure, acute kidney injury, cardiovascular dysfunction, sepsis, trauma, nutrition, and delirium. Results: AI-driven tools demonstrated substantial potential in early detection of clinical deterioration, prediction of outcomes, optimization of mechanical ventilation, identification of acute kidney injury, enhanced cardiovascular monitoring, and improved detection of sepsis and traumatic injuries. AI-assisted imaging systems, including those integrated within PACS, have shown marked improvements in diagnostic accuracy and workflow efficiency. Despite these advancements, significant limitations persist, including data heterogeneity, lack of standardized infrastructures, limited interpretability of algorithmic outputs, risks of bias, and evolving regulatory and ethical considerations. Conclusion: AI has the capacity to augment clinical decision-making, enhance workflow efficiency, and improve patient outcomes in the ICU. However, its real-world impact depends on addressing challenges related to data quality, transparency, fairness, regulatory oversight, and clinician training. With responsible implementation and continued interdisciplinary collaboration, AI is positioned to become an integral component of modern critical care practice. .

Anahtar Kelimeler

• Artificial intelligence
• İntensive care unit
• Machine learning
• Clinical decision support systems

Destekleyen Kurum

This study did not receive support from any specific institution or funding body. All stages of the research, including literature review, analysis, and manuscript preparation, were conducted using the authors’ own academic resources and institutional facilities.

Etik Beyan

As this study is a narrative review based solely on previously published literature, it did not involve human participants, patient data, or animal subjects. Therefore, ethical approval was not required. All included studies were referenced appropriately, and the review adhered to principles of academic integrity and responsible scholarship.

Teşekkür

The author would like to thank their colleagues and institutional library services for providing access to essential literature sources that contributed to the development of this manuscript. The author also acknowledges the broader scientific community whose ongoing research in artificial intelligence and critical care continues to advance the field.

Yoğun Bakımda Yapay Zeka: Uygulamalar, Zorluklar ve Gelecekteki Yönler - Bir İnceleme

Öz

Amaç: Yapay zekâ (YZ), klinisyenlerin hızla gelişen büyük miktarda fizyolojik, laboratuvar ve görüntüleme verisini işlemesi gereken yoğun bakım ünitelerinde (YBÜ) dönüştürücü bir teknoloji olarak ortaya çıkmıştır. Bu derleme, kritik bakımda mevcut YZ uygulamalarını değerlendirmeyi, organ ve sendroma özgü kullanım durumlarını vurgulamayı, önemli uygulama zorluklarını belirlemeyi ve YZ'nin YBÜ uygulamasına güvenli ve etkili bir şekilde entegrasyonu için gerekli gelecekteki yönleri özetlemeyi amaçlamaktadır. Yöntem: Anlatısal bir derleme metodolojisi benimsenmiştir. İlgili literatür, PubMed ve başlıca kritik bakım dergilerinde sistematik olmayan bir arama yoluyla belirlenmiş olup, yakın tarihli klinik, hesaplamalı ve translasyonel çalışmalara odaklanılmıştır. Kanıtlar, tanı, risk sınıflandırması, prognostik modelleme, karar desteği ve görüntüleme analizi gibi fonksiyonel alanlar ve solunum yetmezliği, akut böbrek yetmezliği, kardiyovasküler disfonksiyon, sepsis, travma, beslenme ve deliryum gibi organa özgü uygulamalar genelinde sentezlenmiştir. Sonuçlar: Yapay zekâ destekli araçlar, klinik bozulmanın erken tespiti, sonuçların tahmin edilmesi, mekanik ventilasyonun optimizasyonu, akut böbrek yetmezliğinin belirlenmesi, kardiyovasküler izlemenin iyileştirilmesi ve sepsis ile travmatik yaralanmaların daha iyi tespit edilmesinde önemli bir potansiyel göstermiştir. PACS'e entegre olanlar da dahil olmak üzere yapay zekâ destekli görüntüleme sistemleri, tanısal doğruluk ve iş akışı verimliliğinde belirgin iyileşmeler göstermiştir. Bu gelişmelere rağmen, veri heterojenliği, standartlaştırılmış altyapıların eksikliği, algoritmik çıktıların sınırlı yorumlanabilirliği, önyargı riskleri ve gelişen düzenleyici ve etik hususlar da dahil olmak üzere önemli sınırlamalar devam etmektedir. Sonuç: Yapay zekâ, klinik karar verme süreçlerini destekleme, iş akışı verimliliğini artırma ve yoğun bakım ünitesinde hasta sonuçlarını iyileştirme kapasitesine sahiptir. Bununla birlikte, gerçek dünyadaki etkisi, veri kalitesi, şeffaflık, adalet, düzenleyici denetim ve klinisyen eğitimi ile ilgili zorlukların ele alınmasına bağlıdır. Sorumlu uygulama ve sürekli disiplinler arası işbirliği ile yapay zekâ, modern yoğun bakım uygulamasının ayrılmaz bir bileşeni haline gelmeye hazırdır.

Anahtar Kelimeler

• Yapay Zeka
• Yoğun Bakım Ünitesi
• Makine Öğrenmesi
• Klinik Karar Destek Sistemleri

Destekleyen Kurum

Bu çalışma herhangi bir kurum veya fon kuruluşundan destek almamıştır. Literatür taraması, analiz ve makale hazırlığı da dahil olmak üzere araştırmanın tüm aşamaları, yazarların kendi akademik kaynakları ve kurumsal olanakları kullanılarak gerçekleştirilmiştir.

Etik Beyan

Bu çalışma, yalnızca daha önce yayınlanmış literatüre dayalı bir anlatısal inceleme olduğundan, insan katılımcıları, hasta verileri veya hayvan denekleri içermemiştir. Bu nedenle, etik onay gerekmemiştir. Dahil edilen tüm çalışmalar uygun şekilde referanslandırılmış ve inceleme, akademik dürüstlük ve sorumlu bilimsel çalışma ilkelerine bağlı kalmıştır.

Teşekkür

Yazar, bu makalenin geliştirilmesine katkıda bulunan temel literatür kaynaklarına erişim sağladıkları için meslektaşlarına ve kurum kütüphane hizmetlerine teşekkür eder. Yazar ayrıca, yapay zeka ve yoğun bakım alanındaki devam eden araştırmalarıyla bu alanı geliştirmeye devam eden daha geniş bilim camiasına da minnettar olduğunu belirtir.

Kaynakça

1. Adams R, Henry KE, Sridharan A, Soleimani H, Zhan A, Rawat N, Johnson L, Hager DN, Cosgrove SE, Markowski A, Klein EY, Chen ES, Saheed MO, Henley M, Miranda S, Houston K, Linton RC, Ahluwalia AR, Wu AW, Saria S. (2022). Prospective, multi-site study of patient outcomes after implementation of the TREWS machine learning-based early warning system for sepsis. Nat Med, 28(28(7), 1455–1460.
2. Awad, A., Bader-El-Den, M., McNicholas, J., & Briggs, J. (2017). Early hospital mortality prediction of intensive care unit patients using an ensemble learning approach. International Journal of Medical Informatics, 108, 185-195.
3. Calvert, J., Mao, Q., Hoffman, J. L., Jay, M., Desautels, T., Mohamadlou, H., Chettipally, U., & Das, R. (2016). Using electronic health record collected clinical variables to predict medical intensive care unit mortality. Annals of Medicine & Surgery, 11, 52–57.
4. Choi, D.-J., Park, J. J., Ali, T., & Lee, S. (2020). Artificial intelligence for the diagnosis of heart failure. NPJ Digital Medicine, 3(1), 54.
5. Duron, L., Ducarouge, A., Gillibert, A., Lainé, J., Allouche, C., Cherel, N., Zhang, Z., Nitche, N., Lacave, E., Pourchot, A., Felter, A., Lassalle, L., Regnard, N.-E., & Feydy, A. (2021). Assessment of an AI Aid in Detection of Adult Appendicular Skeletal Fractures by Emergency Physicians and Radiologists: A Multicenter Cross-sectional Diagnostic Study. Radiology, 300(1), 120–129.
6. Ettori, F., Henin, A., Zemmour, C., Chow-Chine, L., Sannini, A., Bisbal, M., Gonzalez, F., Servan, L., De Guibert, J. M., Faucher, M., Boher, J. M., & Mokart, D. (2019). Impact of a computer-assisted decision support system (CDSS) on nutrition management in critically ill hematology patients: The NUTCHOCO study (nutritional care in hematology oncologic patients and critical outcome). Annals of Intensive Care, 9(1), 53.
7. Fagerström, J., Bång, M., Wilhelms, D., & Chew, M. S. (2019). LiSep LSTM: A Machine Learning Algorithm for Early Detection of Septic Shock. Scientific Reports, 9(1), 15132.
8. Gharehchopogh, F. S., & Khalifelu, Z. A. (2011). Neural Network application in diagnosis of patient: A case study. International Conference on Computer Networks and Information Technology, 245–249.

9. Goh, K. H., Wang, L., Yeow, A. Y. K., Poh, H., Li, K., Yeow, J. J. L., & Tan, G. Y. H. (2021). Artificial intelligence in sepsis early prediction and diagnosis using unstructured data in healthcare. Nature Communications, 12(1), 711.
10. Guly, H. R. (2001). Diagnostic errors in an accident and emergency department. Emergency Medicine Journal, 18(4), 263–269.
11. Gutierrez, G. (2020). Artificial Intelligence in the Intensive Care Unit. Critical Care, 24(1), 101.
12. Heaton, J. (2018). Ian Goodfellow, Yoshua Bengio, and Aaron Courville: Deep learning: The MIT Press, 2016, 800 pp, ISBN: 0262035618. Genetic Programming and Evolvable Machines, 19(1–2), 305–307.
13. Hezarjaribi, N., Dutta, R., Xing, T., Murdoch, G. K., Mazrouee, S., Mortazavi, B. J., & Ghasemzadeh, H. (2018). Monitoring Lung Mechanics during Mechanical Ventilation using Machine Learning Algorithms. 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 1160–1163.
14. Kashou, A. H., Noseworthy, P. A., Lopez-Jimenez, F., Attia, Z. I., Kapa, S., Friedman, P. A., & Jentzer, J. C. (2021). The effect of cardiac rhythm on artificial intelligence-enabled ECG evaluation of left ventricular ejection fraction prediction in cardiac intensive care unit patients. International Journal of Cardiology, 339, 54–55.
15. Kim, J., Chae, M., Chang, H.-J., Kim, Y.-A., & Park, E. (2019). Predicting Cardiac Arrest and Respiratory Failure Using Feasible Artificial Intelligence with Simple Trajectories of Patient Data. Journal of Clinical Medicine, 8(9), 1336.
16. Laino, M. E., Ammirabile, A., Lofino, L., Lundon, D. J., Chiti, A., Francone, M., & Savevski, V. (2022). Prognostic findings for ICU admission in patients with COVID-19 pneumonia: Baseline and follow-up chest CT and the added value of artificial intelligence. Emergency Radiology, 29(2), 243–262.
17. Lemeshow, S., Klar, J., & Teres, D. (1995). Outcome prediction for individual intensive care patients: Useful, misused, or abused? Intensive Care Medicine, 21(9), 770–776.
18. Liu, C.-F., Hung, C.-M., Ko, S.-C., Cheng, K.-C., Chao, C.-M., Sung, M.-I., Hsing, S.-C., Wang, J.-J., Chen, C.-J., Lai, C.-C., Chen, C.-M., & Chiu, C.-C. (2022). An artificial intelligence system to predict the optimal timing for mechanical ventilation weaning for intensive care unit patients: A two-stage prediction approach. Frontiers in Medicine, 9, 935366.
19. Lukaszewski, R. A., Yates, A. M., Jackson, M. C., Swingler, K., Scherer, J. M., Simpson, A. J., Sadler, P., McQuillan, P., Titball, R. W., Brooks, T. J. G., & Pearce, M. J. (2008). Presymptomatic Prediction of Sepsis in Intensive Care Unit Patients. Clinical and Vaccine Immunology, 15(7), 1089–1094.
20. Luo, X.-X., Fang, F., So, H.-K., Liu, C., Yam, M.-C., & Lee, A. P.-W. (2017). Automated left heart chamber volumetric assessment using three-dimensional echocardiography in Chinese adolescents. Echo Research and Practice, 4(4), 53–61.
21. Marshall, J. C., Bosco, L., Adhikari, N. K., Connolly, B., Diaz, J. V., Dorman, T., Fowler, R. A., Meyfroidt, G., Nakagawa, S., Pelosi, P., Vincent, J.-L., Vollman, K., & Zimmerman, J. (2017). What is an intensive care unit? A report of the task force of the World Federation of Societies of Intensive and Critical Care Medicine. Journal of Critical Care, 37, 270–276.
22. Mirzakhani, F., Sadoughi, F., Hatami, M., & Amirabadizadeh, A. (2022). Which model is superior in predicting ICU survival: Artificial intelligence versus conventional approaches. BMC Medical Informatics and Decision Making, 22(1), 167.
23. Moazemi, S., Vahdati, S., Li, J., Kalkhoff, S., Castano, L. J. V., Dewitz, B., Bibo, R., Sabouniaghdam, P., Tootooni, M. S., Bundschuh, R. A., Lichtenberg, A., Aubin, H., & Schmid, F. (2023). Artificial intelligence for clinical decision support for monitoring patients in cardiovascular ICUs: A systematic review. Frontiers in Medicine, 10, 1109411.
24. Naqvi, I. H., Mahmood, K., Ziaullaha, S., Kashif, S. M., & Sharif, A. (2016). Better prognostic marker in ICU - APACHE II, SOFA or SAP II! Pakistan Journal of Medical Sciences, 32(5).
25. Nemati, S., Holder, A., Razmi, F., Stanley, M. D., Clifford, G. D., & Buchman, T. G. (2018). An Interpretable Machine Learning Model for Accurate Prediction of Sepsis in the ICU. Critical Care Medicine, 46(4), 547–553.
26. Nguyen, T., Maarek, R., Hermann, A.-L., Kammoun, A., Marchi, A., Khelifi-Touhami, M. R., Collin, M., Jaillard, A., Kompel, A. J., Hayashi, D., Guermazi, A., & Le Pointe, H. D. (2022). Assessment of an artificial intelligence aid for the detection of appendicular skeletal fractures in children and young adults by senior and junior radiologists. Pediatric Radiology, 52(11), 2215–2226.
27. Obermeyer, Z., Powers, B., Vogeli, C., & Mullainathan, S. (2019). Dissecting racial bias in an algorithm used to manage the health of populations. Science, 366(6464), 447–453.
28. Ocagli, H., Bottigliengo, D., Lorenzoni, G., Azzolina, D., Acar, A. S., Sorgato, S., Stivanello, L., Degan, M., & Gregori, D. (2021). A Machine Learning Approach for Investigating Delirium as a Multifactorial Syndrome. International Journal of Environmental Research and Public Health, 18(13), 7105.
29. Ozrazgat-Baslanti, T., Loftus, T. J., Ren, Y., Ruppert, M. M., & Bihorac, A. (2021). Advances in artificial intelligence and deep learning systems in ICU-related acute kidney injury. Current Opinion in Critical Care, 27(6), 560–572.
30. Pérez-Sanpablo, A. I., Quinzaños-Fresnedo, J., Gutiérrez-Martínez, J., Lozano- Rodríguez, I. G., & Roldan-Valadez, E. (2025). Transforming Medical Imaging: The Role of Artificial Intelligence Integration in PACS for Enhanced Diagnostic Accuracy and Workflow Efficiency. Current Medical Imaging Formerly Current Medical Imaging Reviews, 21, e15734056370620.
31. Pimentel, M. A. F., Redfern, O. C., Malycha, J., Meredith, P., Prytherch, D., Briggs, J., Young, J. D., Clifton, D. A., Tarassenko, L., & Watkinson, P. J. (2021). Detecting Deteriorating Patients in the Hospital: Development and Validation of a Novel Scoring System. American Journal of Respiratory and Critical Care Medicine, 204(1), 44–52.
32. Pinsky, M. R., Bedoya, A., Bihorac, A., Celi, L., Churpek, M., Economou-Zavlanos, N. J., Elbers, P., Saria, S., Liu, V., Lyons, P. G., Shickel, B., Toral, P., Tscholl, D., & Clermont, G. (2024). Use of artificial intelligence in critical care: Opportunities and obstacles. Critical Care, 28(1), 113.
33. Ribeiro, M. T., Singh, S., & Guestrin, C. (2016). “Why Should I Trust You?”: Explaining the Predictions of Any Classifier. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 1135–1144.
34. Rieke, N., Hancox, J., Li, W., Milletarì, F., Roth, H. R., Albarqouni, S., Bakas, S., Galtier, M. N., Landman, B. A., Maier-Hein, K., Ourselin, S., Sheller, M., Summers, R. M., Trask, A., Xu, D., Baust, M., & Cardoso, M. J. (2020). The future of digital health with federated learning. NPJ Digital Medicine, 3(1), 119.
35. Schneeberger, D., Stöger, K., & Holzinger, A. (2020). The European Legal Framework for Medical AI. In A. Holzinger, P. Kieseberg, A. M. Tjoa, & E. Weippl (Eds.), Machine Learning and Knowledge Extraction (Vol. 12279, pp. 209–226). Springer International Publishing.
36. Shamout, F. E., Shen, Y., Wu, N., Kaku, A., Park, J., Makino, T., Jastrzębski, S., Witowski, J., Wang, D., Zhang, B., Dogra, S., Cao, M., Razavian, N., Kudlowitz, D., Azour, L., Moore, W., Lui, Y. W., Aphinyanaphongs, Y., Fernandez-Granda, C., & Geras, K. J. (2021). An artificial intelligence system for predicting the deterioration of COVID-19 patients in the emergency department. NPJ Digital Medicine, 4(1), 80.
37. Singh, V., Cheng, S., Kwan, A. C., & Ebinger, J. (2025). United States Food and Drug Administration Regulation of Clinical Software in the Era of Artificial Intelligence and Machine Learning. Mayo Clinic Proceedings: Digital Health, 3(3), 100231.
38. Sinha, P., Delucchi, K. L., McAuley, D. F., O’Kane, C. M., Matthay, M. A., & Calfee, C. S. (2020). Development and validation of parsimonious algorithms to classify acute respiratory distress syndrome phenotypes: A secondary analysis of randomised controlled trials. The Lancet Respiratory Medicine, 8(3), 247–257.
39. Soltan, A. A. S., Yang, J., Pattanshetty, R., Novak, A., Yang, Y., Rohanian, O., Beer, S., Soltan, M. A., Thickett, D. R., Fairhead, R., Zhu, T., Eyre, D. W., Clifton, D. A., Watson, A., Bhargav, A., Tough, A., Rogers, A., Shaikh, A., Valensise, C., … Muthusami, V. (2022). Real-world evaluation of rapid and laboratory-free COVID-19 triage for emergency care: External validation and pilot deployment of artificial intelligence driven screening. The Lancet Digital Health, 4(4), e266–e278.
40. Tomašev, N., Glorot, X., Rae, J. W., Zielinski, M., Askham, H., Saraiva, A., Mottram, A., Meyer, C., Ravuri, S., Protsyuk, I., Connell, A., Hughes, C. O., Karthikesalingam, A., Cornebise, J., Montgomery, H., Rees, G., Laing, C., Baker, C. R., Peterson, K., … Mohamed, S. (2019). A clinically applicable approach to continuous prediction of future acute kidney injury. Nature, 572(7767), 116–119.
41. Wang, D., Li, J., Sun, Y., Ding, X., Zhang, X., Liu, S., Han, B., Wang, H., Duan, X., & Sun, T. (2021). A Machine Learning Model for Accurate Prediction of Sepsis in ICU Patients. Frontiers in Public Health, 9, 754348.
42. Yang, J., Hao, S., Huang, J., Chen, T., Liu, R., Zhang, P., Feng, M., He, Y., Xiao, W., Hong, Y., & Zhang, Z. (2023). The application of artificial intelligence in the management of sepsis. Medical Review, 3(5), 369–380.
43. Yin, W., Yi, Y., Guan, X., Zhou, L., Wang, J., Li, D., & Zuo, X. (2017). Preprocedural Prediction Model for Contrast‐Induced Nephropathy Patients. Journal of the American Heart Association, 6(2), e004498.
44. Yoon, J. H., Pinsky, M. R., & Clermont, G. (2022). Artificial Intelligence in Critical Care Medicine. Critical Care, 26(1), 75.
45. Yu, H., Khalid, M., Touret, A.-S., Bloch, N., Li, B., Qureshi, M. M., Soto, J. A., & Anderson, S. W. (2017). Texture analysis as a radiomic marker for differentiating renal tumors. Abdominal Radiology, 42(10), 2470–2478.