Try a new search

Format these results:

Searched for:

person:castem04

in-biosketch:true

Total Results:

4


Development of a Method for Clinical Evaluation of Artificial Intelligence-Based Digital Wound Assessment Tools

Howell, Raelina S; Liu, Helen H; Khan, Aziz A; Woods, Jon S; Lin, Lawrence J; Saxena, Mayur; Saxena, Harshit; Castellano, Michael; Petrone, Patrizio; Slone, Eric; Chiu, Ernest S; Gillette, Brian M; Gorenstein, Scott A
Importance/UNASSIGNED:Accurate assessment of wound area and percentage of granulation tissue (PGT) are important for optimizing wound care and healing outcomes. Artificial intelligence (AI)-based wound assessment tools have the potential to improve the accuracy and consistency of wound area and PGT measurement, while improving efficiency of wound care workflows. Objective/UNASSIGNED:To develop a quantitative and qualitative method to evaluate AI-based wound assessment tools compared with expert human assessments. Design, Setting, and Participants/UNASSIGNED:This diagnostic study was performed across 2 independent wound centers using deidentified wound photographs collected for routine care (site 1, 110 photographs taken between May 1 and 31, 2018; site 2, 89 photographs taken between January 1 and December 31, 2019). Digital wound photographs of patients were selected chronologically from the electronic medical records from the general population of patients visiting the wound centers. For inclusion in the study, the complete wound edge and a ruler were required to be visible; circumferential ulcers were specifically excluded. Four wound specialists (2 per site) and an AI-based wound assessment service independently traced wound area and granulation tissue. Main Outcomes and Measures/UNASSIGNED:The quantitative performance of AI tracings was evaluated by statistically comparing error measure distributions between test AI traces and reference human traces (AI vs human) with error distributions between independent traces by 2 humans (human vs human). Quantitative outcomes included statistically significant differences in error measures of false-negative area (FNA), false-positive area (FPA), and absolute relative error (ARE) between AI vs human and human vs human comparisons of wound area and granulation tissue tracings. Six masked attending physician reviewers (3 per site) viewed randomized area tracings for AI and human annotators and qualitatively assessed them. Qualitative outcomes included statistically significant difference in the absolute difference between AI-based PGT measurements and mean reviewer visual PGT estimates compared with PGT estimate variability measures (ie, range, standard deviation) across reviewers. Results/UNASSIGNED:A total of 199 photographs were selected for the study across both sites; mean (SD) patient age was 64 (18) years (range, 17-95 years) and 127 (63.8%) were women. The comparisons of AI vs human with human vs human for FPA and ARE were not statistically significant. AI vs human FNA was slightly elevated compared with human vs human FNA (median [IQR], 7.7% [2.7%-21.2%] vs 5.7% [1.6%-14.9%]; P < .001), indicating that AI traces tended to slightly underestimate the human reference wound boundaries compared with human test traces. Two of 6 reviewers had a statistically higher frequency in agreement that human tracings met the standard area definition, but overall agreement was moderate (352 yes responses of 583 total responses [60.4%] for AI and 793 yes responses of 1166 total responses [68.0%] for human tracings). AI PGT measurements fell in the typical range of variation in interreviewer visual PGT estimates; however, visual PGT estimates varied considerably (mean range, 34.8%; mean SD, 19.6%). Conclusions and Relevance/UNASSIGNED:This study provides a framework for evaluating AI-based digital wound assessment tools that can be extended to automated measurements of other wound features or adapted to evaluate other AI-based digital image diagnostic tools. As AI-based wound assessment tools become more common across wound care settings, it will be important to rigorously validate their performance in helping clinicians obtain accurate wound assessments to guide clinical care.
PMID: 34009348
ISSN: 2574-3805
CID: 4877232

Hyperbaric oxygen therapy for COVID-19 patients with respiratory distress: treated cases versus propensity-matched controls

Gorenstein, Scott A; Castellano, Michael L; Slone, Eric S; Gillette, Brian; Liu, Helen; Alsamarraie, Cindy; Jacobson, Alan M; Wall, Stephen P; Adhikari, Samrachana; Swartz, Jordan L; McMullen, Jenica J S; Osorio, Marcela; Koziatek, Christian A; Lee, David C
Objective/UNASSIGNED:Given the high mortality and prolonged duration of mechanical ventilation of COVID-19 patients, we evaluated the safety and efficacy of hyperbaric oxygen for COVID-19 patients with respiratory distress. Methods/UNASSIGNED:This is a single-center clinical trial of COVID-19 patients at NYU Winthrop Hospital from March 31 to April 28, 2020. Patients in this trial received hyperbaric oxygen therapy at 2.0 atmospheres of pressure in monoplace hyperbaric chambers for 90 minutes daily for a maximum of five total treatments. Controls were identified using propensity score matching among COVID-19 patients admitted during the same time period. Using competing-risks survival regression, we analyzed our primary outcome of inpatient mortality and secondary outcome of mechanical ventilation. Results/UNASSIGNED:We treated 20 COVID-19 patients with hyperbaric oxygen. Ages ranged from 30 to 79 years with an oxygen requirement ranging from 2 to 15 liters on hospital days 0 to 14. Of these 20 patients, two (10%) were intubated and died, and none remain hospitalized. Among 60 propensity-matched controls based on age, sex, body mass index, coronary artery disease, troponin, D-dimer, hospital day, and oxygen requirement, 18 (30%) were intubated, 13 (22%) have died, and three (5%) remain hospitalized (with one still requiring mechanical ventilation). Assuming no further deaths among controls, we estimate that the adjusted subdistribution hazard ratios were 0.37 for inpatient mortality (p=0.14) and 0.26 for mechanical ventilation (p=0.046). Conclusion/UNASSIGNED:Though limited by its study design, our results demonstrate the safety of hyperbaric oxygen among COVID-19 patients and strongly suggests the need for a well-designed, multicenter randomized control trial.
PMID: 32931666
ISSN: 1066-2936
CID: 4591182

A Wolf in Sheep's Clothing: An Unusual Presentation of Diabetic Myonecrosis

Boinpally, Harika; Howell, Raelina S; Mazzie, Joseph; Slone, Eric; Woods, Jon S; Gillette, Brian M; Castellano, Michael; Gorenstein, Scott
GENERAL PURPOSE/UNASSIGNED:To provide information about the diagnosis and treatment of diabetic myonecrosis (DMN).This continuing education activity is intended for physicians, physician assistants, nurse practitioners, and nurses with an interest in skin and wound care.After participating in this educational activity, the participant should be better able to:1. Cite the incidence and symptomatology of diabetic myonecrosis.2. Identify the diagnostic tests associated with DMN.3. Summarize the evidence-based treatments for DMN.Diabetic myonecrosis is a rare complication of poorly controlled diabetes mellitus that presents similarly to many common conditions such as cellulitis, abscess, and fasciitis. Therefore, a high index of suspicion is required for diagnosis. Magnetic resonance imaging is the investigative test of choice. Treatment includes antiplatelet therapy, nonsteroidal anti-inflammatory agents, and glycemic control.
PMID: 30134275
ISSN: 1538-8654
CID: 3260982

Wound Care Center of Excellence: Guide to Operative Technique for Chronic Wounds

Howell, Raelina S; Gorenstein, Scott; Castellano, Michael; Slone, Eric; Woods, Jon S; Gillette, Brian M; Donovan, Virginia; Criscitelli, Theresa; Brem, Harold; Brathwaite, Collin
PMID: 29154922
ISSN: 1879-1190
CID: 2986042