scholarly journals A physiologically realistic virtual patient database for the study of arterial haemodynamics

2021 ◽  
Author(s):  
G. Jones ◽  
J. Parr ◽  
P. Nithiarasu ◽  
S. Pant
Keyword(s):  
Virtual Patient ◽  
Patient Database ◽  
Arterial Haemodynamics

scholarly journals Artificial intelligence approaches to predict coronary stenosis severity using non-invasive fractional flow reserve

2020 ◽  
Vol 234 (11) ◽  
pp. 1337-1350 ◽  
Author(s):  
Jason M Carson ◽  
Neeraj Kavan Chakshu ◽  
Igor Sazonov ◽  
Perumal Nithiarasu
Keyword(s):  
Neural Network ◽  
Fractional Flow Reserve ◽  
Virtual Patient ◽  
Fractional Flow ◽  
Feed Forward Neural Network ◽  
Feed Forward ◽  
Single Vessel ◽  
Flow Reserve ◽  
Patient Database ◽  
Real Patients

Fractional flow reserve is the current reference standard in the assessment of the functional impact of a stenosis in coronary heart disease. In this study, three models of artificial intelligence of varying degrees of complexity were compared to fractional flow reserve measurements. The three models are the multivariate polynomial regression, which is a statistical method used primarily for correlation; the feed-forward neural network; and the long short-term memory, which is a type of recurrent neural network that is suited to modelling sequences. The models were initially trained using a virtual patient database that was generated from a validated one-dimensional physics-based model. The feed-forward neural network performed the best for all test cases considered, which were a single vessel case from a virtual patient database, a multi-vessel network from a virtual patient database, and 25 clinically invasive fractional flow reserve measurements from real patients. The feed-forward neural network model achieved around 99% diagnostic accuracy in both tests involving virtual patients, and a respectable 72% diagnostic accuracy when compared to the invasive fractional flow reserve measurements. The multivariate polynomial regression model performed well in the single vessel case, but struggled on network cases as the variation of input features was much larger. The long short-term memory performed well for the single vessel cases, but tended to have a bias towards a positive fractional flow reserve prediction for the virtual multi-vessel case, and for the patient cases. Overall, the feed-forward neural network shows promise in successfully predicting fractional flow reserve in real patients, and could be a viable option if trained using a large enough data set of real patients.

scholarly journals Machine learning for detection of stenoses and aneurysms: application in a physiologically realistic virtual patient database

2021 ◽  
Author(s):  
G. Jones ◽  
J. Parr ◽  
P. Nithiarasu ◽  
S. Pant
Keyword(s):  
Machine Learning ◽  
Arterial Disease ◽  
Virtual Patient ◽  
Aortic Aneurysms ◽  
Artery Stenosis ◽  
Arterial System ◽  
Gradient Boosting ◽  
High Severity ◽  
Patient Database ◽  
The Common

AbstractThis study presents an application of machine learning (ML) methods for detecting the presence of stenoses and aneurysms in the human arterial system. Four major forms of arterial disease—carotid artery stenosis (CAS), subclavian artery stenosis (SAS), peripheral arterial disease (PAD), and abdominal aortic aneurysms (AAA)—are considered. The ML methods are trained and tested on a physiologically realistic virtual patient database (VPD) containing 28,868 healthy subjects, adapted from the authors previous work and augmented to include disease. It is found that the tree-based methods of Random Forest and Gradient Boosting outperform other approaches. The performance of ML methods is quantified through the $$F_1$$ F 1 score and computation of sensitivities and specificities. When using six haemodynamic measurements (pressure in the common carotid, brachial, and radial arteries; and flow-rate in the common carotid, brachial, and femoral arteries), it is found that maximum $$F_1$$ F 1 scores larger than 0.9 are achieved for CAS and PAD, larger than 0.85 for SAS, and larger than 0.98 for both low- and high-severity AAAs. Corresponding sensitivities and specificities are larger than 90% for CAS and PAD, larger than 85% for SAS, and larger than 98% for both low- and high-severity AAAs. When reducing the number of measurements, performance is degraded by less than 5% when three measurements are used, and less than 10% when only two measurements are used for classification. For AAA, it is shown that $$F_1$$ F 1 scores larger than 0.85 and corresponding sensitivities and specificities larger than 85% are achievable when using only a single measurement. The results are encouraging to pursue AAA monitoring and screening through wearable devices which can reliably measure pressure or flow-rates.

Virtual Patient Encounters

2010 ◽  
Vol 43 (5) ◽  
pp. 62
Author(s):  
CHRIS NOTTE ◽  
NEIL SKOLNIK
Keyword(s):  
Virtual Patient

Virtual Patient Encounters

2010 ◽  
Vol 40 (4) ◽  
pp. 53
Author(s):  
CHRIS NOTTE ◽  
NEIL SKOLNIK
Keyword(s):  
Virtual Patient

MS2/Cardio: Towards a Multi-Service Medical Software for Cardiology

1992 ◽  
Vol 31 (01) ◽  
pp. 18-28 ◽  
Author(s):  
C. Combi ◽  
G. Pozzi ◽  
R. Rossi ◽  
F. Pinciroli
Keyword(s):  
Statistical Processing ◽  
Clinical Information ◽  
Daily Practice ◽  
Management Unit ◽  
Record Management ◽  
Medical Software ◽  
Software Packages ◽  
Ambulatory Patients ◽  
Patient Database ◽  
Individual Program

Abstract:Many clinics are interested to use software packages in daily practice, but lack of integration of such packages seriously limits their scope. In practice this often entails switching between programs and interrupting the run of an individual program. A multi-task approach would not solve this problem as it would not eliminate the need to input the same data many times, as often occurs when using separate packages. The construction of a Multi-Service Medical Software package (MSx2) is described, which was also developed as an example of practical integration of some clinically relevant functions. The package runs on a personal computer in an MS-DOS environment and integrates a time-oriented medical record management unit (TOMRU) for data of ambulatory patients, and a drug information management unit (DIMU) concerning posology, content, effects, and possible interactions. Of the possible database configurations allowed by MSx2, the cardiology patient database (MSx2/C) and hypertensive patient database (MSx2/H) were developed and described here. Clinical information to be included in the configurations was obtained after discussion and consensus of clinical practitioners. MSx2/C was distributed to several hundred clinical centers during computerized courses to train future users. MSx2 can easily transfer patient data to statistical processing packages.

scholarly journals Does the shoe-lace technique aid direct closure of fasciotomy wounds after acute compartment syndrome of the lower leg? a retrospective case-control study

2021 ◽  
pp. 145749692110196
Author(s):  
P. Suomalainen ◽  
T.-K. Pakarinen ◽  
I. Pajamäki ◽  
M. K. Laitinen ◽  
H.-J. Laine ◽  
...  
Keyword(s):  
Compartment Syndrome ◽  
Outcome Measurement ◽  
Controlled Trial ◽  
Lower Leg ◽  
Conventional Technique ◽  
Tibia Fracture ◽  
Acute Compartment Syndrome ◽  
Retrospective Case ◽  
Direct Closure ◽  
Patient Database

Background & aim: Tibia fractures are relatively common injuries that are accompanied with acute compartment syndrome in approximately 2% to 20% of cases. Although the shoe-lace technique, where vessel loops are threaded in a crisscross fashion and tightened daily, has been widely used, no studies have compared the shoe-lace technique with the conventional one. The aim of this study was to compare the shoe-lace technique with the conventional technique. Materials and Methods: We identified 359 consecutive patients with intramedullary nailed tibia fracture and complete medical records including outpatient data between April 2007 and April 2015 from electronic patient database of our institute. The use of the shoe-lace technique was compared to conventional one (in which wounds were first left open with moist dressings). Main outcome measurement is direct closure of fasciotomy wounds. Results: From 359 consecutive patients with intramedullary nailed tibia fracture, fasciotomy was performed on 68 (19%) patients. Of these, the shoe-lace technique was used in 47 (69%) patients while in 21 (31%) patients, the shoe-lace technique was not applied. Side-to-side approximation was successful in 36 patients (77%) in the shoe-lace+ group and 7 patients (33%) in the shoe-lace– group (p = 0.002). Conclusion: The main finding of our comparative study was that the shoe-lace technique seems to ease direct closure of lower leg fasciotomy wounds, and thus reduces the frequency of free skin grafts. Our finding needs to be confirmed in a high-quality randomized controlled trial.

Demonstration: Virtual Patient Immersive Trainer to Train Perceptual Skills Using Augmented Reality

2020 ◽  
Vol 64 (1) ◽  
pp. 470-472
Author(s):  
Christen E. Sushereba ◽  
Laura G. Militello
Keyword(s):  
Medical Students ◽  
Augmented Reality ◽  
Student Performance ◽  
Group Interaction ◽  
Virtual Patient ◽  
Perceptual Cues ◽  
Perceptual Skills ◽  
Individual Interaction ◽  
Severity Of Injury ◽  
Microsoft Hololens

In this session, we will demonstrate the Virtual Patient Immersive Trainer (VPIT). The VPIT system uses augmented reality (AR) to allow medics and medical students to experience a photorealistic, life-sized virtual patient. The VPIT supports learners in obtaining the perceptual skills required to recognize and interpret subtle perceptual cues critical to assessing a patient’s condition. We will conduct an interactive demonstration of the virtual patient using both a tablet (for group interaction) and an AR-enabled headset (Microsoft HoloLens) for individual interaction. In addition, we will demonstrate use of the instructor tablet to control what the learner sees (e.g., injury types, severity of injury) and to monitor student performance.

Sign in / Sign up

Export Citation Format

Share Document