Medical Diagnosis from Images with Intuitionistic Fuzzy Distance Measures
| Autor: | Roan Thi Ngan, Le Hoang Son, Tran Manh Tuan, Bui Cong Cuong |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
Measure (data warehouse)
Similarity (geometry) 020205 medical informatics Degree (graph theory) business.industry Computer science Intuitionistic fuzzy 02 engineering and technology Machine learning computer.software_genre Distance measures New diagnosis Image (mathematics) 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Artificial intelligence Medical diagnosis business computer |
| Zdroj: | Rough Sets ISBN: 9783319993676 IJCSR |
| Popis: | Medical diagnosis from images supports clinicians in their profession. In practical dentistry, diseases are found mainly on experience of dentists regarding dental structures and explicit symptoms of patients. In this paper, in order to reduce errors in medical diagnosis problem from images, we introduce a new diagnostic model based on intuitionistic fuzzy distance measures with parameter learning. A new intuitionistic fuzzy distance measure named Modified H-max is proposed to calculate similarity degree between an input image and all patterns of corresponding disease patterns. Parameters of the proposed measure are trained to optimize performance. Hence, the new diagnosis model has the advantages of using the cross-evaluation degree of H-max measure and weight optimization. The proposed algorithm is experimentally validated on real datasets of Hanoi Medical University, Vietnam against related methods. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|