Novel Noninvasive Brain Disease Detection System Using a Facial Image Sensor
Brain disease including any conditions or disabilities that affect the brain is fast becoming a leading cause of death. The traditional diagnostic methods of brain disease are time-consuming, inconvenient and non-patient friendly. As more and more individuals undergo examinations to determine if the...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2017-12-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/17/12/2843 |
| _version_ | 1828284000353910784 |
|---|---|
| author | Ting Shu Bob Zhang Yuan Yan Tang |
| author_facet | Ting Shu Bob Zhang Yuan Yan Tang |
| author_sort | Ting Shu |
| collection | DOAJ |
| description | Brain disease including any conditions or disabilities that affect the brain is fast becoming a leading cause of death. The traditional diagnostic methods of brain disease are time-consuming, inconvenient and non-patient friendly. As more and more individuals undergo examinations to determine if they suffer from any form of brain disease, developing noninvasive, efficient, and patient friendly detection systems will be beneficial. Therefore, in this paper, we propose a novel noninvasive brain disease detection system based on the analysis of facial colors. The system consists of four components. A facial image is first captured through a specialized sensor, where four facial key blocks are next located automatically from the various facial regions. Color features are extracted from each block to form a feature vector for classification via the Probabilistic Collaborative based Classifier. To thoroughly test the system and its performance, seven facial key block combinations were experimented. The best result was achieved using the second facial key block, where it showed that the Probabilistic Collaborative based Classifier is the most suitable. The overall performance of the proposed system achieves an accuracy −95%, a sensitivity −94.33%, a specificity −95.67%, and an average processing time (for one sample) of <1 min at brain disease detection. |
| first_indexed | 2024-04-13T08:53:58Z |
| format | Article |
| id | doaj.art-0818b35d01ad4a05a4a14f337a220b2f |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-04-13T08:53:58Z |
| publishDate | 2017-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-0818b35d01ad4a05a4a14f337a220b2f2022-12-22T02:53:22ZengMDPI AGSensors1424-82202017-12-011712284310.3390/s17122843s17122843Novel Noninvasive Brain Disease Detection System Using a Facial Image SensorTing Shu0Bob Zhang1Yuan Yan Tang2Department of Computer and Information Science, Avenida da Universidade, University of Macau, Taipa, Macau 999078, ChinaDepartment of Computer and Information Science, Avenida da Universidade, University of Macau, Taipa, Macau 999078, ChinaDepartment of Computer and Information Science, Avenida da Universidade, University of Macau, Taipa, Macau 999078, ChinaBrain disease including any conditions or disabilities that affect the brain is fast becoming a leading cause of death. The traditional diagnostic methods of brain disease are time-consuming, inconvenient and non-patient friendly. As more and more individuals undergo examinations to determine if they suffer from any form of brain disease, developing noninvasive, efficient, and patient friendly detection systems will be beneficial. Therefore, in this paper, we propose a novel noninvasive brain disease detection system based on the analysis of facial colors. The system consists of four components. A facial image is first captured through a specialized sensor, where four facial key blocks are next located automatically from the various facial regions. Color features are extracted from each block to form a feature vector for classification via the Probabilistic Collaborative based Classifier. To thoroughly test the system and its performance, seven facial key block combinations were experimented. The best result was achieved using the second facial key block, where it showed that the Probabilistic Collaborative based Classifier is the most suitable. The overall performance of the proposed system achieves an accuracy −95%, a sensitivity −94.33%, a specificity −95.67%, and an average processing time (for one sample) of <1 min at brain disease detection.https://www.mdpi.com/1424-8220/17/12/2843image sensorbrain diseasenoninvasive detection systemfacial key block analysisProCRCmedical biometrics |
| spellingShingle | Ting Shu Bob Zhang Yuan Yan Tang Novel Noninvasive Brain Disease Detection System Using a Facial Image Sensor Sensors image sensor brain disease noninvasive detection system facial key block analysis ProCRC medical biometrics |
| title | Novel Noninvasive Brain Disease Detection System Using a Facial Image Sensor |
| title_full | Novel Noninvasive Brain Disease Detection System Using a Facial Image Sensor |
| title_fullStr | Novel Noninvasive Brain Disease Detection System Using a Facial Image Sensor |
| title_full_unstemmed | Novel Noninvasive Brain Disease Detection System Using a Facial Image Sensor |
| title_short | Novel Noninvasive Brain Disease Detection System Using a Facial Image Sensor |
| title_sort | novel noninvasive brain disease detection system using a facial image sensor |
| topic | image sensor brain disease noninvasive detection system facial key block analysis ProCRC medical biometrics |
| url | https://www.mdpi.com/1424-8220/17/12/2843 |
| work_keys_str_mv | AT tingshu novelnoninvasivebraindiseasedetectionsystemusingafacialimagesensor AT bobzhang novelnoninvasivebraindiseasedetectionsystemusingafacialimagesensor AT yuanyantang novelnoninvasivebraindiseasedetectionsystemusingafacialimagesensor |