Currently, a tension exists between proposed research criteria for determining symptom severity and clinical operationalization of the same. Proposed DSM-5 autism spectrum criteria includes three severity classifications: Level 1 Mild / (“Requiring support”), Level 2 Moderate (“Requiring substantial support”), and Level 3 severe (“Requiring very substantial support”) (American Psychiatric Association 2012). These classifications are split across two areas, Social Communication (SC) and Restricted and Repetitive Behaviors (RRB ) . Since fuzz has a status of Certainty , can a machine learning algorithm classify it. For example : (train data with gradually decrease until get accuracy for level 1 which Mild then train again until level 2 Social Communication (SC). Do i need to decouple two method (Fuzzy and machine learning technique)?
Based upon sufficient data availability, you may choose booth of these or you may use booth simultaneously. For your initial review you may consider these readings : https://www.nature.com/articles/tp201210
https://pure.hud.ac.uk/ws/files/16870701/Health_systems_Article.pdf
Article A new machine learning model based on induction of rules for...
if i exactly understood, you must perform multiclass classification: on 3 classes (level 1, level 2, level 3) or 2 classes (SC or RRB). If you need machine learning approaches and fuzzy logic, you can use neural fuzzy nets, which has a lot of advantages against neural networks and fuzzy logic, and you can training it. But you must have dataset.
Digital pathology has entered a new era with the availability of whole slide scanners that create the high-resolution images of full biopsy slides. Consequently, the uncertainty regarding the correspondence between the image areas and the diagnostic labels assigned by pathologists at the slide level, and the need for identifying regions that belong to multiple classes with different clinical significances have emerged as two new challenges. However, generalizability of the state-of-the-art algorithms, whose accuracies were reported on carefully selected regions of interest (ROIs) for the binary benign versus cancer classification, to these multi-class learning and localization problems is currently unknown.
- Badges
- Science topic