Improving CNN training on endoscopic image data by extracting additionally training data from endoscopic videos
نویسندگان
چکیده
منابع مشابه
Endoscopic Sedation: From Training to Performance
Adequate sedation and analgesia are considered essential requirements to relieve patient discomfort and pain and ultimately to improve the outcomes of modern gastrointestinal endoscopic procedures. The willingness of patients to undergo sedation during endoscopy has increased steadily in recent years and standard sedation practices are needed for both patient safety and successful procedural ou...
متن کاملTraining in Endoscopy: Endoscopic Ultrasound
Endoscopic ultrasound (EUS) has been recently established as an indispensable modality for the diagnosis and management of pancreatobiliary and gastrointestinal (GI) disorders. EUS proficiency requires both cognitive and technical abilities, including an understanding of the appropriate indications, the performance of appropriate evaluations before and after the procedure, and the management of...
متن کاملTechnique training: endoscopic percutaneous tracheostomy.
Percutaneous tracheostomy is being used increasingly in the intensive care unit and endoscopic control of this procedure affords an improved level of safety. Training in such new minimal access techniques can be a significant risk factor in patient outcome. Surgical simulation provides training which minimizes this risk. We present a method of training in percutaneous endoscopic tracheostomy us...
متن کاملImproving ASR performance on PDA by contamination of training data
Automatic Speech Recognition (ASR) on Personal Digital Assistant (PDA) suffers from the intrinsic hardware characteristics of the audio interface, for example, low quality microphones and device internal noises. In this paper, we propose to compensate for these weaknesses by contaminating clean training data with the distortion sources that are specific to the target device. We present a method...
متن کاملFast Training of Support Vector Machines by Extracting Boundary Data
Support vector machines have gotten wide acceptance for their high generalization ability for real world applications. But the major drawback is slow training for classification problems with a large number of training data. To overcome this problem, in this paper, we discuss extracting boundary data from the training data and train the support vector machine using only these data. Namely, for ...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Computerized Medical Imaging and Graphics
سال: 2020
ISSN: 0895-6111
DOI: 10.1016/j.compmedimag.2020.101798