Heuristic Weight Initialization for Diagnosing Heart Diseases Using Feature Ranking
The advent of Artificial Intelligence (AI) has had a broad impact on life to solve various tasks. Building AI models and integrating them with modern technologies is a central challenge for researchers. These technologies include wearables and implants in living beings, and their use is known as hum...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-10-01
|
| Series: | Technologies |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7080/11/5/138 |
| _version_ | 1827719797167620096 |
|---|---|
| author | Musulmon Lolaev Shraddha M. Naik Anand Paul Abdellah Chehri |
| author_facet | Musulmon Lolaev Shraddha M. Naik Anand Paul Abdellah Chehri |
| author_sort | Musulmon Lolaev |
| collection | DOAJ |
| description | The advent of Artificial Intelligence (AI) has had a broad impact on life to solve various tasks. Building AI models and integrating them with modern technologies is a central challenge for researchers. These technologies include wearables and implants in living beings, and their use is known as human augmentation, using technology to enhance human abilities. Combining human augmentation with artificial intelligence (AI), especially after the recent successes of the latter, is the most significant advancement in their applicability. In the first section, we briefly introduce these modern applications in health care and examples of their use cases. Then, we present a computationally efficient AI-driven method to diagnose heart failure events by leveraging actual heart failure data. The classifier model is designed without conventional models such as gradient descent. Instead, a heuristic is used to discover the optimal parameters of a linear model. An analysis of the proposed model shows that it achieves an accuracy of 84% and an <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>F</mi><mn>1</mn></msub></semantics></math></inline-formula> score of 0.72 with only one feature. With five features for diagnosis, the accuracy achieved is 83%, and the <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>F</mi><mn>1</mn></msub></semantics></math></inline-formula> score is 0.74. Moreover, the model is flexible, allowing experts to determine which variables are more important than others when implementing diagnostic systems. |
| first_indexed | 2024-03-10T20:51:07Z |
| format | Article |
| id | doaj.art-8db9e06a58934a0491adb74ef292eebf |
| institution | Directory Open Access Journal |
| issn | 2227-7080 |
| language | English |
| last_indexed | 2024-03-10T20:51:07Z |
| publishDate | 2023-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Technologies |
| spelling | doaj.art-8db9e06a58934a0491adb74ef292eebf2023-11-19T18:20:26ZengMDPI AGTechnologies2227-70802023-10-0111513810.3390/technologies11050138Heuristic Weight Initialization for Diagnosing Heart Diseases Using Feature RankingMusulmon Lolaev0Shraddha M. Naik1Anand Paul2Abdellah Chehri3The School of Computer Science and Engineering, Kyungpook National University, Daegu 41566, Republic of KoreaThe School of Computer Science and Engineering, Kyungpook National University, Daegu 41566, Republic of KoreaThe School of Computer Science and Engineering, Kyungpook National University, Daegu 41566, Republic of KoreaDepartment of Mathematics and Computer Science, Royal Military College of Canada, Kingston, ON 17000, CanadaThe advent of Artificial Intelligence (AI) has had a broad impact on life to solve various tasks. Building AI models and integrating them with modern technologies is a central challenge for researchers. These technologies include wearables and implants in living beings, and their use is known as human augmentation, using technology to enhance human abilities. Combining human augmentation with artificial intelligence (AI), especially after the recent successes of the latter, is the most significant advancement in their applicability. In the first section, we briefly introduce these modern applications in health care and examples of their use cases. Then, we present a computationally efficient AI-driven method to diagnose heart failure events by leveraging actual heart failure data. The classifier model is designed without conventional models such as gradient descent. Instead, a heuristic is used to discover the optimal parameters of a linear model. An analysis of the proposed model shows that it achieves an accuracy of 84% and an <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>F</mi><mn>1</mn></msub></semantics></math></inline-formula> score of 0.72 with only one feature. With five features for diagnosis, the accuracy achieved is 83%, and the <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>F</mi><mn>1</mn></msub></semantics></math></inline-formula> score is 0.74. Moreover, the model is flexible, allowing experts to determine which variables are more important than others when implementing diagnostic systems.https://www.mdpi.com/2227-7080/11/5/138Artificial Intelliencedata mininggradient decentfeature selectionheart disease |
| spellingShingle | Musulmon Lolaev Shraddha M. Naik Anand Paul Abdellah Chehri Heuristic Weight Initialization for Diagnosing Heart Diseases Using Feature Ranking Technologies Artificial Intellience data mining gradient decent feature selection heart disease |
| title | Heuristic Weight Initialization for Diagnosing Heart Diseases Using Feature Ranking |
| title_full | Heuristic Weight Initialization for Diagnosing Heart Diseases Using Feature Ranking |
| title_fullStr | Heuristic Weight Initialization for Diagnosing Heart Diseases Using Feature Ranking |
| title_full_unstemmed | Heuristic Weight Initialization for Diagnosing Heart Diseases Using Feature Ranking |
| title_short | Heuristic Weight Initialization for Diagnosing Heart Diseases Using Feature Ranking |
| title_sort | heuristic weight initialization for diagnosing heart diseases using feature ranking |
| topic | Artificial Intellience data mining gradient decent feature selection heart disease |
| url | https://www.mdpi.com/2227-7080/11/5/138 |
| work_keys_str_mv | AT musulmonlolaev heuristicweightinitializationfordiagnosingheartdiseasesusingfeatureranking AT shraddhamnaik heuristicweightinitializationfordiagnosingheartdiseasesusingfeatureranking AT anandpaul heuristicweightinitializationfordiagnosingheartdiseasesusingfeatureranking AT abdellahchehri heuristicweightinitializationfordiagnosingheartdiseasesusingfeatureranking |