Eye Segmentation Method for Telehealth: Application to the Myasthenia Gravis Physical Examination.

Due to the precautions put in place during the COVID-19 pandemic, utilization of telemedicine has increased quickly for patient care and clinical trials. Unfortunately, teleconsultation is closer to a video conference than a medical consultation, with the current solutions setting the patient and doctor into an evaluation that relies entirely on a two-dimensional view of each other. We are developing a patented telehealth platform that assists with diagnostic testing of ocular manifestations of myasthenia gravis. We present a hybrid algorithm combining deep learning with computer vision to give quantitative metrics of ptosis and ocular muscle fatigue leading to eyelid droop and diplopia. The method works both on a fixed image and frame by frame of the video in real-time, allowing capture of dynamic muscular weakness during the examination. We then use signal processing and filtering to derive robust metrics of ptosis and l ocular misalignment. In our construction, we have prioritized the robustness of the method versus accuracy obtained in controlled conditions in order to provide a method that can operate in standard telehealth conditions. The approach is general and can be applied to many disorders of ocular motility and ptosis.

The SmartOR: a distributed sensor network to improve operating room efficiency.

BACKGROUND: Despite the significant expense of OR time, best practice achieves only 70% efficiency. Compounding this problem is a lack of real-time data. Most current OR utilization programs require manual data entry. Automated systems require installation and maintenance of expensive tracking hardware throughout the institution. This study developed an inexpensive, automated OR utilization system and analyzed data from multiple operating rooms. STUDY DESIGN: OR activity was deconstructed into four room states. A sensor network was then developed to automatically capture these states using only three sensors, a local wireless network, and a data capture computer. Two systems were then installed into two ORs, recordings captured 24/7. The SmartOR recorded the following events: any room activity, patient entry/exit time, anesthesia time, laparoscopy time, room turnover time, and time of preoperative patient identification by the surgeon. RESULTS: From November 2014 to December 2015, data on 1003 cases were collected. The mean turnover time was 36 min, and 38% of cases met the institutional goal of ≤30 min. Data analysis also identified outlier cases (>1 SD from mean) in the domains of time from patient entry into the OR to intubation (11% of cases) and time from extubation to patient exiting the OR (11% of cases). Time from surgeon identification of patient to scheduled procedure start time was 11 min (institution bylaws require 20 min before scheduled start time), yet OR teams required 22 min on average to bring a patient into the room after surgeon identification. CONCLUSION: The SmartOR automatically and reliably captures data on OR room state and, in real time, identifies outlier cases that may be examined closer to improve efficiency. As no manual entry is required, the data are indisputable and allow OR teams to maintain a patient-centric focus.

A robust and non-obtrusive automatic event tracking system for operating room management to improve patient care.

BACKGROUND: Optimization of OR management is a complex problem as each OR has different procedures throughout the day inevitably resulting in scheduling delays, variations in time durations and overall suboptimal performance. There exists a need for a system that automatically tracks procedural progress in real time in the OR. This would allow for efficient monitoring of operating room states and target sources of inefficiency and points of improvement. STUDY DESIGN: We placed three wireless sensors (floor-mounted pressure sensor, ventilator-mounted bellows motion sensor and ambient light detector, and a general room motion detector) in two ORs at our institution and tracked cases 24 h a day for over 4 months. RESULTS: We collected data on 238 total cases (107 laparoscopic cases). A total of 176 turnover times were also captured, and we found that the average turnover time between cases was 35 min while the institutional goal was 30 min. Deeper examination showed that 38 % of laparoscopic cases had some aspect of suboptimal activity with the time between extubation and patient exiting the OR being the biggest contributor (16 %). CONCLUSION: Our automated system allows for robust, wireless real-time OR monitoring as well as data collection and retrospective data analyses. We plan to continue expanding our system and to project the data in real time for all OR personnel to see. At the same time, we plan on adding key pieces of technology such as RFID and other radio-frequency systems to track patients and physicians to further increase efficiency and patient safety.

A Digital Telehealth System to Compute the Myasthenia Gravis Core Examination Metrics.

Background: Telemedicine practice for neurological diseases has grown significantly during the COVID-19 pandemic.Telemedicine offers an opportunity to assess digitalization of examinations and enhances access to modern computer vision and artificial intelligence processing to annotate and quantify examinations in a consistent and reproducible manner. The Myasthenia Gravis Core Examination (MG-CE) has been recommended for the telemedicine evaluation of patients with myasthenia gravis. Objective: We aimed to assess the ability to take accurate and robust measurements during the examination, which would allow improvement in workflow efficiency by making the data acquisition and analytics fully automatic and thereby limit the potential for observation bias. Methods: We used Zoom (Zoom Video Communications) videos of patients with myasthenia gravis undergoing the MG-CE. The core examination tests required 2 broad categories of processing. First, computer vision algorithms were used to analyze videos with a focus on eye or body motions. Second, for the assessment of examinations involving vocalization, a different category of signal processing methods was required. In this way, we provide an algorithm toolbox to assist clinicians with the MG-CE. We used a data set of 6 patients recorded during 2 sessions. Results: Digitalization and control of quality of the core examination are advantageous and let the medical examiner concentrate on the patient instead of managing the logistics of the test. This approach showed the possibility of standardized data acquisition during telehealth sessions and provided real-time feedback on the quality of the metrics the medical doctor is assessing. Overall, our new telehealth platform showed submillimeter accuracy for ptosis and eye motion. In addition, the method showed good results in monitoring muscle weakness, demonstrating that continuous analysis is likely superior to pre-exercise and post-exercise subjective assessment. Conclusions: We demonstrated the ability to objectively quantitate the MG-CE. Our results indicate that the MG-CE should be revisited to consider some of the new metrics that our algorithm identified. We provide a proof of concept involving the MG-CE, but the method and tools developed can be applied to many neurological disorders and have great potential to improve clinical care.

A Systems Approach to Assess Transport and Diffusion of Hazardous Airborne Particles in a Large Surgical Suite: Potential Impacts on Viral Airborne Transmission.

Airborne transmission of viruses, such as the coronavirus 2 (SARS-CoV-2), in hospital systems are under debate: it has been shown that transmission of SARS-CoV-2 virus goes beyond droplet dynamics that is limited to 1 to 2 m, but it is unclear if the airborne viral load is significant enough to ensure transmission of the disease. Surgical smoke can act as a carrier for tissue particles, viruses, and bacteria. To quantify airborne transmission from a physical point of view, we consider surgical smoke produced by thermal destruction of tissue during the use of electrosurgical instruments as a marker of airborne particle diffusion-transportation. Surgical smoke plumes are also known to be dangerous for human health, especially to surgical staff who receive long-term exposure over the years. There are limited quantified metrics reported on long-term effects of surgical smoke on staff's health. The purpose of this paper is to provide a mathematical framework and experimental protocol to assess the transport and diffusion of hazardous airborne particles in every large operating room suite. Measurements from a network of air quality sensors gathered during a clinical study provide validation for the main part of the model. Overall, the model estimates staff exposure to airborne contamination from surgical smoke and biological material. To address the clinical implication over a long period of time, the systems approach is built upon previous work on multi-scale modeling of surgical flow in a large operating room suite and takes into account human behavior factors.

Gastroenterology Procedures Generate Aerosols: An Air Quality Turnover Solution to Mitigate COVID-19's Propagation Risk.

The growing fear of virus transmission during the 2019 coronavirus disease (COVID-19) pandemic has called for many scientists to look into the various vehicles of infection, including the potential to travel through aerosols. Few have looked into the issue that gastrointestinal (GI) procedures may produce an abundance of aerosols. The current process of risk management for clinics is to follow a clinic-specific HVAC formula, which is typically calculated once a year and assumes perfect mixing of the air within the space, to determine how many minutes each procedural room refreshes 99% of its air between procedures when doors are closed. This formula is not designed to fit the complex dynamic of small airborne particle transport and deposition that can potentially carry the virus in clinical conditions. It results in reduced procedure throughput as well as an excess of idle time in clinics that process a large number of short procedures such as outpatient GI centers. We present and tested a new cyber-physical system that continuously monitors airborne particle counts in procedural rooms and also at the same time automatically monitors the procedural rooms' state and flexible endoscope status without interfering with the clinic's workflow. We use our data gathered from over 1500 GI cases in one clinical suite to understand the correlation between air quality and standard procedure types as well as identify the risks involved with any HVAC system in a clinical suite environment. Thanks to this system, we demonstrate that standard GI procedures generate large quantities of aerosols, which can potentially promote viral airborne transmission among patients and healthcare staff. We provide a solution for the clinic to improve procedure turnover times and throughput, as well as to mitigate the risk of airborne transmission of the virus.

A model of workflow in the hospital during a pandemic to assist management.

We present a computational model of workflow in the hospital during a pandemic. The objective is to assist management in anticipating the load of each care unit, such as the ICU, or ordering supplies, such as personal protective equipment, but also to retrieve key parameters that measure the performance of the health system facing a new crisis. The model was fitted with good accuracy to France's data set that gives information on hospitalized patients and is provided online by the French government. The goal of this work is both practical in offering hospital management a tool to deal with the present crisis of COVID-19 and offering a conceptual illustration of the benefit of computational science during a pandemic.