Predicting Outcomes in Patients Undergoing Pancreatectomy Using Wearable Technology and Machine Learning: Prospective Cohort Study

Abstract

Background Pancreatic cancer is the third leading cause of cancer-related deaths, and although pancreatectomy is currently the only curative treatment, it is associated with significant morbidity. Objective The objective of this study was to evaluate the utility of wearable telemonitoring technologies to predict treatment outcomes using patient activity metrics and machine learning. Methods In this prospective, single-center, single-cohort study, patients scheduled for pancreatectomy were provided with a wearable telemonitoring device to be worn prior to surgery. Patient clinical data were collected and all patients were evaluated using the American College of Surgeons National Surgical Quality Improvement Program surgical risk calculator (ACS-NSQIP SRC). Machine learning models were developed to predict whether patients would have a textbook outcome and compared with the ACS-NSQIP SRC using area under the receiver operating characteristic (AUROC) curves. Results Between February 2019 and February 2020, 48 patients completed the study. Patient activity metrics were collected over an average of 27.8 days before surgery. Patients took an average of 4162.1 (SD 4052.6) steps per day and had an average heart rate of 75.6 (SD 14.8) beats per minute. Twenty-eight (58%) patients had a textbook outcome after pancreatectomy. The group of 20 (42%) patients who did not have a textbook outcome included 14 patients with severe complications and 11 patients requiring readmission. The ACS-NSQIP SRC had an AUROC curve of 0.6333 to predict failure to achieve a textbook outcome, while our model combining patient clinical characteristics and patient activity data achieved the highest performance with an AUROC curve of 0.7875. Conclusions Machine learning models outperformed ACS-NSQIP SRC estimates in predicting textbook outcomes after pancreatectomy. The highest performance was observed when machine learning models incorporated patient clinical characteristics and activity metrics.

The Application and Comparison of Machine Learning Models for the Prediction of Breast Cancer Prognosis: Retrospective Cohort Study

“…Compared with the Cox proportional hazard regression, machine learning methods do not make any parametric or semiparametric assumptions and have the ability to detect and account for higher-order interactions as well as nonlinear relationships [8]. While there have been some attempts to use machine learning to build cancer prognosis prediction models [6,[9][10][11][12][13], currently, there is no consensus on whether traditional or machine learning-based prognostic prediction models have a better predictive performance.…”

A Smarter Health through the Internet of Surgical Things

“…In a 2021 study by Cos et al, patients scheduled to undergo pancreatic surgery were monitored preoperatively by using a wearable smart device that was able to record heart rate, activity status, etc., and through an internet connection, transmit them to a central server. Not only did patients adhere to this novel concept, but the data that were automatically collected were of such quality that the research team developed an accurate predictive model for postoperative outcomes [3]. Biosensor-based systems are able to wirelessly transmit data on physiological parameters of the patients in order to assist with postoperative monitoring.…”

“…The utilization of the Internet and IoT in medical practice can take many shapes and forms. Ranging from the awe-inspiring telesurgical procedures [1,2] to complex AI machine learning applications that aid in medical decision making [3], to a simple email containing a preoperative CT scan, the Internet of Surgical Things (IoST) is here to stay. A representative example of the IoST is a smart ingestible sensor (pill) that is activated after being swallowed [4], "travels" in the body through the colon and sends data to outer devices such as computers and smartphones when it detects a threat for cancer.…”

“…Such a device can be used instead of colonoscopy for people who cannot obtain a colonoscopy due to psychological and physiological problems. In the broader aspect of the IoST context, IoST entities are different types of connected entities that "live" in this smart setting, such as surgical things (e.g., a connected surgery tool) [5], organs (e.g., a connected colon, an artificial heart) [6], humans (e.g., a connected patient or doctor) [7], smart devices [8] (e.g., a connected heartbeat monitoring device, a smart pill, an ingestible sensor), services [3] (e.g., a connected telemonitoring service), data (e.g., a connected data steam of heart monitoring data), etc. In this extensive survey, we aim to present an overview of current uses of IoT-embedded surgical practice by focusing on Telesurgery and Surgical Telementoring.…”

“…There is, however, a lack of a systematic approach within the existing literature as to how the IoT concept has revolutionized the surgical world towards smarter health. To fill this gap in current knowledge, the present systematic literature review studies the Internet of Surgical Things (IoST) domain by focusing on the three most prominent areas of application: (a) image-guided surgery, (b) telesurgery and telementoring in surgery, and (c) surgical patient monitoring [3,5]. Between these three areas of applications within the surgical discipline, the applications of Telesurgery and Surgical Telementoring are undoubtedly the most influenced by the IoMT concept.…”