A Roadmap for Automatic Surgical Site Infection Detection and Evaluation Using User-Generated Incision Images

Abstract

Background: Emerging technologies such as smartphones and wearable sensors have enabled the paradigm shift to new patient-centered healthcare, together with recent mobile health (mHealth) app development. One such promising healthcare app is incision monitoring based on patient-taken incision images. In this review, challenges and potential solution strategies are investigated for surgical site infection (SSI) detection and evaluation using surgical site images taken at home. Methods: Potential image quality issues, feature extraction, and surgical site image analysis challenges are discussed. Recent image analysis and machine learning solutions are reviewed to extract meaningful representations as image markers for incision monitoring. Discussions on opportunities and challenges of applying these methods to derive accurate SSI prediction are provided. Conclusions: Interactive image acquisition as well as customized image analysis and machine learning methods for SSI monitoring will play critical roles in developing sustainable mHealth apps to achieve the expected outcomes of patient-taken incision images for effective out-of-clinic patient-centered healthcare with substantially reduced cost.

Telemedicine and the COVID-19 Pandemic: Are We Ready to Go Live?

“…Healthcare systems have only begun to scratch the surface of telemedicine's benefits. The potential for technological advancement in the field is unbounded 30,31 as medical devices and smartphones push the limits of what services can be delivered remotely (see the next section). 32…”

Optical Biopsy Using a Neural Network to Predict Gene Expression From Photos of Wounds

“…To our knowledge, this is the first study that has attempted to predict gene expression in situ wounds from digital photographs using ML. While there has been fairly extensive prior work on the use of ML methods to characterize wounds, these approaches focus on classification tasks such as wound type 15 or the presence and/or absence of infection, [16][17][18][19] or segmentation to aid in quantifying wound size 20 , 21 ; none of these projects have attempted to gain insight into the mechanistic and/or functional state of the wounds. The closest-related work involves the application of a CNN to infer gene expression profiles from whole-slide images used for the histologic analysis of tumors 29 ; this work aimed to identify particular markers for tumor behavior presumably based on specific patterns of cellular subtypes as reflecting certain genetic instructions.…”

“…and segmentation (e.g., where in this image is a tumor?). In fact, these methods have already been applied to photographs of wounds, intending to aid in the classification of wounds by type 15 or presence of infection [16][17][18][19] (classification tasks) and size and/or edge determination 20 , 21 (segmentation tasks). However, despite their potential clinical utility, none of this prior work has aimed to provide granular mechanistic insight into the state of the wound, and these methods are therefore limited both in terms of providing more granular mechanistic sub-stratification of wounds (that would aid in the design and execution of a clinical trial) or aiding the translation of mechanistic knowledge regarding wound biology into the development of novel therapeutics.…”

Current and future use of telemedicine in surgical clinics during and beyond COVID-19: A narrative review

“…This protocol improved post-operative follow-up decreasing the need for patients to attend ambulatory care for routine wound checks [ 12 ]. Machine learning and virtual reality are also being developed to assist with post-operative wound assessment [ 16 ].…”

“…A range of telemedicine modalities can be used separately, or in unison, by the patient and healthcare provider to optimise communication levels whilst located remotely from one another[ 12 , 13 ]. Telemedicine encompasses telephone, videoconferencing interaction5, engagement of consumers remotely via advanced medical tablets in a ‘virtual exam room’[ 14 ], through smartphone SMS and application technology[ [15] , [16] , [17] ], via established social media networks [ 18 ] or even access real-time diagnostic elements (such as remote video-otoscopy19). In surgical clinics, an appropriate means to substitute for key points of assessment is essential due to the interventional nature of surgery.…”