Preoperative Risk Prediction of Postoperative Complications for Elective Cardiac Surgery using At-home Smartphone Dynamometry

Project Summary/Abstract Surgical complication represents one of the biggest factors impacting quality of care and is one of the major burdens on the healthcare system. There are 40-50 million surgical procedures performed in the US annually with a mortality rate of 1.3% (650,000 people) and morbidity rate of ~14% (7M people). The mortality rates

significantly increase for patients who are frail, with a heightened risk of 5.1%, and continues to increase into 90 and 180 days for those who are deemed very frail, reaching a staggering 43%. The difficulty in increasing surgical success is not necessarily in improving the surgical procedures, but rather the perioperative care that surrounds

the surgery: pre-habilitating the patient into fitness prior to surgery, improving patient recovery to discharge patients to recover comfortably at-home, detecting onsets of complications early to provide non-emergent treatment. The main objective of this project is to develop a hand grip strength (HGS) measurement solution

based completely on a smartphone application that converts the phone’s vibration motor and sensor into a mobile dynamometer. Our scientific premise, demonstrated by a berth of clinical evidence, is that hand grip strength provides a biomarker of physical frailty that corresponds to general physiologic reserve and cardiopulmonary

status, as well as systemic inflammation. Prior studies have linked a diminutive HGS preoperatively to heightened risk of surgical complications. This corresponds well to evidence that frailty, which correlates strongly to a weakened HGS, is a major risk factor for surgical complications due to the extremely taxing nature of surgical

procedures on the body requiring a level of fitness to recover after the operation. Although HGS as a measure is possible with commercial HGS dynamometers, a smartphone-sensor enabled digital dynamometer can make measurements guided through adaptive interface, automatically digitized, and integrated with ML analytics. With

the patient’s own smartphone, progress of pre-habilitation can be assessed to determine fitness to undergo surgery, changes in health status can be detected, and timely changes in surgical plan can be made. To increase the scalability of HGS screening, we propose a smartphone assessment that patients, including older adults, can

administer themselves at home that tracks changes in HGS during the preoperative period. We will further develop and evaluate different machine learning algorithms that use the HGS feature biomarkers measured by the phone to perform automated risk prediction of postsurgical outcomes. Because the project would be carried

out in the rich research context of UC San Diego Division of Perioperative Informatics in the School of Medicine in conjunction with the Anesthesiology Preparedness Clinic, it will be possible to validate the mobile dynamometer assessments with a cohort of surgical patients undergoing medium to high-risk procedures that

would benefit most from a stratified risk assessment protocol. Our translational goal is to provide access to low- cost, digital at-home HGS monitoring to transform the way perioperative care is conducted, from one that is sparse and poorly assessed, to one that is highly integrated through close monitoring and feedback.