Deep into Laboratory: An Artificial Intelligence Approach to Recommend Laboratory Tests

Abstract

Laboratory tests are performed to make effective clinical decisions. However, inappropriate laboratory test ordering hampers patient care and increases financial burden for healthcare. An automated laboratory test recommendation system can provide rapid and appropriate test selection, potentially improving the workflow to help physicians spend more time treating patients. The main objective of this study was to develop a deep learning-based automated system to recommend appropriate laboratory tests. A retrospective data collection was performed at the National Health Insurance database between 1 January 2013, and 31 December 2013. We included all prescriptions that had at least one laboratory test. A total of 1,463,837 prescriptions from 530,050 unique patients was included in our study. Of these patients, 296,541 were women (55.95%), the range of age was between 1 and 107 years. The deep learning (DL) model achieved a higher area under the receiver operating characteristics curve (AUROC micro = 0.98, and AUROC macro = 0.94). The findings of this study show that the DL model can accurately and efficiently identify laboratory tests. This model can be integrated into existing workflows to reduce under- and over-utilization problems.

Laboratory Demand Management Strategies—An Overview

“…Currently, few published articles deal with the issue of applying AI algorithms to laboratory test selection. Islam et al [ 56 , 57 ] have published two such studies, one of which in this issue, where they developed a deep learning algorithm based on retrospective patient data to predict appropriate laboratory tests. Xu et al [ 58 ] aimed to identify superfluous tests in existing lab orders by estimating normal test results within a retrospective dataset.…”

“…Furthermore, it has to be acknowledged that AI is only a tool of assistance [ 57 ]. A combination of computerized and physician-guided processes may be better than each one on their own.…”

Artificial intelligence and thyroid disease management

“…Guiding the physicians for the ordering of the right TFT according to clinical context is important, especially in a context where the over-use of TFT has been documented in both hospital and primary care practices ( 15 ). The optimal choice of TFT is important to make effective clinical decision, to help physicians spend more time treating patients ( 16 ). In contrast, excessive TFT ordering policies can prompt financial weight in a period of rising medical care costs ( 8 ).…”

“…Artificial intelligence based companions have the potential to help physicians to optimize TFT prescription and to define intelligent order sets that can contribute to reduce laboratory overutilization ( 15 ). A recent study investigated the value of deep learning-based automated system to recommend appropriate laboratory tests ( 16 ). The AI based model achieved a higher area under the receiver operating characteristics curve (AUROC micro = 0.98, and AUROC macro = 0.94).…”

Rise of the Machines: The Inevitable Evolution of Medicine and Medical Laboratories Intertwining with Artificial Intelligence—A Narrative Review

“…In this issue the same authors present a similar DL model for personalized laboratory test prediction, which was developed using 1,463,837 lab orders from 530,050 unique patients [44]. This model achieved an AUROC of 0.92-0.96 for 114 laboratory tests, 0.96-1 for 106 laboratory tests, and 0.88-0.92, 0.84-0.88, 0.80-0.84 and 0.76-0.80 for 56, 30, 5, and 4 laboratory tests, respectively.…”