When AI Is Not Enough: Reducing Diagnostic Errors with Radiologist Oversight

References

• Argon N, Ziya S (2009) Priority assignment under imperfect information on customer type identities. Manufacturing Service Oper. Management 11(4):674–693.Link, Google Scholar
• Ariel D, Susan P (2024) Aidoc: Building a hospital-centric AI platform. Harvard Business School Case 624-046 1–24, Cambridge, MA.Google Scholar
• Atar R, Giat C, Shimkin N (2010) The cμ/θ rule for many-server queues with abandonment. Oper. Res. 58(5):1427–1439.Link, Google Scholar
• Atar R, Mandelbaum A, Reiman M (2004) Scheduling a multi class queue with many exponential servers: Asymptotic optimality in heavy traffic. Ann. Appl. Probab. 14(3):1084–1134.Crossref, Google Scholar
• Baltruschat I, Steinmeister L, Nickisch H, Saalbach A, Grass M, Adam G, Knopp T, Ittrich H (2021) Smart chest X-ray worklist prioritization using artificial intelligence: A clinical workflow simulation. Eur. Radiology 31(6):3837–3845.Crossref, Google Scholar
• Barron Y, Baron O (2022) On dedicated versus pooled service in the presence of triage errors. Working paper, Ariel University, Ariel, Israel.Google Scholar
• Batra K, Xi Y, Bhagwat S, Espino A, Peshock R (2023) Radiologist worklist reprioritization using artificial intelligence: Impact on report turnaround times for CTPA examinations positive for acute pulmonary embolism. Amer. J. Roentgenology 221(3):324–333.Crossref, Google Scholar
• Cox D, Smith W (1961) Queues (Methuen, London).Google Scholar
• Dai T, Abràmoff M (2023) Incorporating artificial intelligence into healthcare workflows: Models and insights. Tutorials in Operations Research: Advancing the Frontiers of OR/MS: From Methodologies to Applications (INFORMS, Catonsville, MD), 133–155.Link, Google Scholar
• Dai T, Singh S (2024) Using AI as gatekeeper or second opinion: Designing patient pathways for AI-augmented healthcare. Working paper, Carey Business School, Johns Hopkins University, Baltimore.Google Scholar
• Dai T, Tayur S (2022) Designing AI-augmented healthcare delivery systems for physician buy-in and patient acceptance. Production Oper. Management 31(12):4443–4451.Crossref, Google Scholar
• Gaube S, Suresh H, Raue M, Merritt A, Berkowitz S, Lermer E, Coughlin J, Guttag JV, Colak E, Ghassemi M (2021) Do as AI say: Susceptibility in deployment of clinical decision-aids. NPJ Digital Medicine 4(1):31.Crossref, Google Scholar
• Harrison J, Zeevi A (2004) Dynamic scheduling of a multiclass queue in the Halfin-Whitt heavy traffic regime. Oper. Res. 52(2):243–257.Link, Google Scholar
• He C, Liu W, Xu J, Huang Y, Dong Z, Wu Y, Kharrazi H (2024) Efficiency, accuracy, and health professional’s perspectives regarding artificial intelligence in radiology practice: A scoping review. iRadiology 2(2):156–172.Crossref, Google Scholar
• Hosny A, Parmar C, Quackenbush J, Schwartz L, Aerts H (2018) Artificial intelligence in radiology. Nature Rev. Cancer 18(8):500–510.Crossref, Google Scholar
• Hu Y, Chan C, Dong J (2022) Optimal scheduling of proactive service with customer deterioration and improvement. Management Sci. 68(4):2533–2578.Link, Google Scholar
• Huang J, Carmeli B, Mandelbaum A (2015) Control of patient flow in emergency departments, or multiclass queues with deadlines and feedback. Oper. Res. 63(4):892–908.Link, Google Scholar
• Ibanez M, Clark J, Huckman R, Staats B (2018) Discretionary task ordering: Queue management in radiological services. Management Sci. 64(9):4389–4407.Link, Google Scholar
• Kamali M, Tezcan T, Yildiz O (2019) When to use provider triage in emergency departments. Management Sci. 65(3):1003–1019.Link, Google Scholar
• Lee C, Nagy P, Weaver S, Newman-Toker D (2013) Cognitive and system factors contributing to diagnostic errors in radiology. Amer. J. Roentgenology 201(3):611–617.Crossref, Google Scholar
• Li D, Shen F, Yin Y, Peng J, Chen P (2013) Weighted Youden index and its two-independent-sample comparison based on weighted sensitivity and specificity. Chinese Medical J. 126(6):1150–1154.Crossref, Google Scholar
• Lidal I, Holte H, Vist G (2013) Triage systems for pre-hospital emergency medical services—A systematic review. Scandinavian J. Trauma Resuscitation Emergency Medicine 21:1–6.Crossref, Google Scholar
• Little J (1961) A proof for the queuing formula: L=λW. Oper. Res. 9(3):383–387.Link, Google Scholar
• Long Z, Shimkin N, Zhang H, Zhang J (2020) Dynamic scheduling of multiclass many-server queues with abandonment: The generalized cμ/h rule. Oper. Res. 68(4):1218–1230.Link, Google Scholar
• Mandelbaum A, Stolyar A (2004) Scheduling flexible servers with convex delay costs: Heavy-traffic optimality of the generalized cμ-rule. Oper. Res. 52(6):836–855.Link, Google Scholar
• Monaghan T, Rahman S, Agudelo C, Wein A, Lazar J, Everaert K, Dmochowski R (2021) Foundational statistical principles in medical research: Sensitivity, specificity, positive predictive value, and negative predictive value. Medicina 57(5):503.Crossref, Google Scholar
• Papadimitriou C, Tsitsiklis J (1999) The complexity of optimal queuing network control. Math. Oper. Res. 24(2):293–305.Link, Google Scholar
• Saghafian S, Hopp W, Iravani S, Cheng Y, Diermeier D (2018) Workload management in telemedical physician triage and other knowledge-based service systems. Management Sci. 64(11):5180–5197.Link, Google Scholar
• Saghafian S, Hopp WJ, Van Oyen MP, Desmond JS, Kronick SL (2014) Complexity-augmented triage: A tool for improving patient safety and operational efficiency. Manufacturing Service Oper. Management 16(3):329–345.Link, Google Scholar
• Singh S, Van Mieghem J, Gurvich I (2024) Feature-driven priority queuing. Working paper, Leavey School of Business, Santa Clara University, Santa Clara, CA.Google Scholar
• Sun Z, Argon N, Ziya S (2018) Patient triage and prioritization under austere conditions. Management Sci. 64(10):4471–4489.Link, Google Scholar
• Sun Z, Argon N, Ziya S (2022) When to triage in service systems with hidden customer class identities? Production Oper. Management 31(1):172–193.Crossref, Google Scholar
• Van der Zee S, Theil H (1961) Priority assignment in waiting-line problems under conditions of misclassification. Oper. Res. 9(6):875–885.Link, Google Scholar
• Van Mieghem J (1995) Dynamic scheduling with convex delay costs: The generalized cμ rule. Ann. Appl. Probab. 5(3):809–833.Crossref, Google Scholar
• Whitt W (2002) Stochastic-process limits: An introduction to stochastic-process limits and their application to queues. Space 500:391–426.Google Scholar
• Zhong Z, Cao P, Huang J, Zhou S (2024) Capacity allocation and scheduling in two-stage service systems with multiclass customers. Manufacturing Service Oper. Management 26(5):1842–1859.Link, Google Scholar
• Zychlinski N (2023) Applications of fluid models in service operations management. Queueing Systems 103(1):161–185.Crossref, Google Scholar
• Zychlinski N (2024) Managing queues with reentrant customers in support of hybrid healthcare. Stochastic Systems 14(2):167–190.Link, Google Scholar