Balance Scan: Automation of the Timed Up and Go
Boca Raton, FL November 2022
- In initial evaluation of the CaptureProof TUG software Machine Learning Automated Algorithm, the longest TUG time was for a subject aged 60 with TUG time of 52 seconds; the longest TUG time in the 65 and over category was 42 seconds. Agreement between human frame-by-frame analysis versus the ML automated algorithm was 91% for TUG times under 15 seconds and 77% for all cases. The agreement between the stopwatch and frame-by-frame analysis was 59%, and 56% between stopwatch and ML.
- The major source of error was attributed to lose hospital gowns, which reduced the precision of detecting the human form. More of the subjects with shorter TUG times wore street clothing which tended to be tighter than hospital gowns. The video record permits establishing trends over time and indicate progress or regress. One parameter derived by ML is the time from the second turn (turn before sitting) to the time the subject is seated. Several subjects rushed to the chair and essentially rolled into the seat. Their turn-to-sit time was short < 0.7 seconds. Despite “safe” TUG their walk was not steady. Additional parameters are yet to be evaluated. These include listing and hip and knee positions.
- All ages 37% passed the TUG test with a TUG time of < 11 seconds, a category considered low risk. 13% of the subjects were at moderate risk, with TUG times between 11 - 13 seconds, 50% of the subjects were over 13 seconds. Over 65 28% were at low or moderate risk of falling. Assisted devices (walkers and canes) accounted for 20% of subjects; 88% had TUG times over 20 seconds. But not all assisted device patients were at risk despite the long TUG time based on medical evaluation.
Jerry WInniczek PhD1, Rhonda Taubin, MD2, Joseph Bledsoe MD2, Kevin Christiansen OT2, William Beninati MD2, Eric Liston2, Meghan Conroy1 Organizations: 1CAPTUREPROOF, Inc, 2Intermountain Healthcare
Video quality using outpatient smartphone videos in epilepsy: Results from the OSmartViE study
European Journal of Neurology JANUARY 2021
- Forty‐four patients (31 women, age 45.1 years [r = 20–82]) were included and 530 SVs were viewed by a mean of seven experts and six residents; one video per patient was reviewed for a mean of 133.8 s (r = 9–543).
- In all, 30 patients had PNEAs, 11 had ESs, and three had PhysNEEs.
- Quality was suitable in 70.8% of SVs (375/530 total views), with 36/44 (81.8%) patient SVs rated as adequate by the majority of reviewers.
- Accuracy improved with the presence of convulsive features from 72.4% to 98.2% in ESs and from 71.1% to 95.7% in PNEAs.
- An accurate diagnosis was given by all reviewers (100%) in 11/44 SVs (all PNEAs).
- Audio was rated as good by 86.2% of reviewers for these SVs compared with 75.4% for the remaining SVs (p = 0.01).
- Lighting was better in SVs associated with high accuracy (p = 0.06), but clarity was not (p = 0.59).
- Poor video quality yielded unknown diagnoses in 24.2% of the SVs reviewed.
- Features hindering diagnosis were limited interactivity, restricted field of view and short video duration.
William O. Tatum, Lawrence J. Hirsch, Michael A. Gelfand, Emily K. Acton, W. Curt LaFrance, Robert B. Duckrow, David Chen, Andrew S. Blum, John Hixson, Joe Drazkowski, Selim Benbadis, Gregory D. Cascino, The OSmartViE Collaborators
ASSESSMENT OF THE PREDICTIVE VALUE OF OUTPATIENT SMARTPHONE VIDEOS FOR DIAGNOSIS OF EPILEPTIC SEIZURES
JAMA NEUROLOGY JANUARY 2020
- The odds of receiving a correct diagnosis were 5.45 times greater using smartphone video alongside patient history and physical examination results than with history and physical examination alone (95% CI, 1.01-54.3; P = .02)
- One-fourth of the smartphone videos were correctly diagnosed by 100% of the reviewing physicians, composed solely of psychogenic attacks.
- When histories and physical examination results were combined with smartphone videos, correct diagnoses rose from 78.6% to 95.2%.
William O. Tatum, DO1; Lawrence J. Hirsch, MD2; Michael A. Gelfand, MD3; Emily K. Acton, BS3; W. Curt LaFrance Jr, MD, MPH4; Robert B. Duckrow, MD2; David K. Chen, MD5; Andrew S. Blum, MD4; John D. Hixson, MD6; Joe F. Drazkowski, MD7; Selim R. Benbadis, MD8; Gregory D. Cascino, MD9; for the OSmartViE Investigators
1Department of Neurology, Mayo Clinic, Jacksonville, Florida
2Department of Neurology, Yale University, New Haven, Connecticut
3Department of Neurology, University of Pennsylvania, Philadelphia
4Department of Neurology, Brown University, Providence, Rhode Island
5Department of Neurology, Baylor College of Medicine, Houston, Texas
6University of California, San Francisco, San Francisco
7Department of Neurology, Mayo Clinic, Phoenix, Arizona
8Department of Neurology, University of South Florida, Tampa
9Department of Neurology, Mayo Clinic, Rochester, Minnesota
Noninvasive assessment of liver steatosis in deceased donors: A pilot study
JOURNAL OF LIVER TRANSPLANTATION DECEMBER 2017
Manuela Cesaretti, Nicolas Poté, Francois Cauchy, Federica Dondero, Safi Dokmak, Ailton Sepulveda, Anne Sophie Schneck, Claire Francoz, Francois Durand, Valerie Paradis and Olivier Soubrane PARIS, FRANCE
Outpatient Smartphone Videos for Classifying Epileptic and Nonepileptic Seizures
PRESENTED AT AMERICAN EPILEPSY SOCIETY MEETING 2017
William O. Tatum DO1, Larry Hirsch MD2, Robert Duckrow MD2, David Chen MD3, Michael Gelfand MD PhD4, Curt LaFrance MD5, Andrew Blum MD5, John Hixson MD6, Joe Drazkowski MD1, Selim Benbadis MD7, Diego Carvalho MD1, Alfonso Lopez MD1, Erin Okazaki MD1, Iris Marin Collazo MD1, Ashish Ranpura MD2, Scott Yuan MD2, Jon Kleen MD6, Erin Coonan8, Gregory Cascino MD1
Departments of Neurology: 1Mayo Clinic, 2Yale University, 3Baylor University, 4University of Pennsylvania, 5Brown University, 6University of California San Francisco, 7University of South Florida, 8Boston College
Outpatient Smartphone Videos in Epilepsy (OSmartViE): Initial Results of Video Quality
POSTER PRESENTED AT AMERICAN EPILEPSY SOCIETY MEETING DEC 2017
Erin E. Coonan1, Lawrence J. Hirsch MD2, Robert B. Duckrow MD2, David Chen MD3, Michael Gelfand MD PhD4, Andrew Blum MD5, John Hixson MD6, William LaFrance MD5, Joseph Drazkowski MD1, Selim Benbadis MD7, Gregory Cascino MD1, William O. Tatum DO1
Departments of Neurology: 1Mayo Clinic, 2Yale University, 3Baylor University, 4University of Pennsylvania, 5Brown University, 6University of California San Francisco, 7University of South Florida
Noninvasive assessment of liver steatosis in deceased donors: a pilot study
POSTER PRESENTED AT 13éme Congrès Francophone de Chirurgie Digestive et Hépato-Bilio-Pancréatique 13th Congress of Digestive Surgery and Hepato Pancreatic Bilio
Manuela Cesaretti1 2, Nicolas Poté3, Francois Cauchy1, Federica Dondero1, Safi Dokmak1, Ailton Sepulveda1, Anne Sophie Schneck1, Claire Francoz4, Francois Durand4, Valerie Paradis3, Olivier Soubrane1
1Beaujon Hospital, Department: HPB Surgery and Liver Transplantation; 2Istituto Italiano di Tecnologia; 3Hôpital Beaujon, Department: Pathology; 4Beaujon Hospital, Hepatology and Liver Intensive Care.
Outpatient Smartphone Videos in Epilepsy (OSmartViE): Initial Results
POSTER PRESENTED AT AMERICAN EPILEPSY SOCIETY DEC 2016
William O. Tatum DO1, Larry Hirsch MD2, Robert Duckrow MD2, David Chen MD3, Michael Gelfand MD PhD4, Curt LaFrance MD5, Andrew Blum MD5, John Hixson MD6, Joe Drazkowski MD1, Selim Benbadis MD7, Diego Carvalho MD1, Alfonso Lopez MD1, Erin Okazaki MD1, Iris Marin Collazo MD1, Ashish Ranpura MD2, Scott Yuan MD2, Jon Kleen MD6, Erin Coonan8, Gregory Cascino MD1 Departments of Neurology: 1Mayo Clinic, 2Yale University, 3Baylor University, 4University of Pennsylvania, 5Brown University, 6University of California San Francisco, 7University of South Florida, 8Boston University
TESTING feasibility and utility of remote data capture technology to assess parkinson’s disease
POSTER PRESENTED AT INTERNATIONAL PARKINSON AND MOVEMENT DISORDER SOCIETY BERLIN, GERMANY JUNE 2016
- High quality of home video recordings for asynchronous video-recordings are feasible in early to mid-stage PD using a HIPAA-compliant app and cloud-based platform
- Ratings from the modified “svUPDRS” are in good agreement with scores on equivalent items from the in- person UPDRS
- Two independent raters did not differ significantly in ratings on the video-based visits
- 1 hour in-office training is adequate for most patients
- Asynchronous virtual visits are a novel tool for creating an archive of rich patient data beyond the scope & scale of what can be captured in a clinical setting
CLAIRE HENCHCLIFFE, MD DPhil, cornell medical center, new york, ny
REMOTE ASYNCHRONOUS TELEREHABILITATION FOLLOWING TOTAL KNEE ARTHROPLASTY
JOURNAL OF TELEMEDICINE TELECARE MARCH 2016
STEFANO BINI, MD, KAISER PERMANENTE, OAKLAND, CA
Use of Smartphone Application for Management of Postoperative Wound Complications
Krzysztof B. Siemionow, MD, UNIVERSITY ILLINOIS CHICAGO, CHICAGO, IL
BENEFITS OF MEDICAL MEDIA IN PEDIATRIC NEUROLOGY
AWARDED “ORAL PRESENTATION” AT AAN POSTER SESSION APRIL 2013
FARHARD SAHEBKAR-MOGHADDAM, MD, SUTTER MEDICAL CENTER, SAN FRANCISCO, CA