Enhancing Cancer Diagnosis
Early diagnosis of cancer is crucial for improved patient results. With the aim of improving the effectiveness of cancer diagnosis, this paper introduces a new proposed method, computer-aided diagnosis, utilizing the level-set algorithm based on the edge detection approach for medical image segment...
Saved in:
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Koya University
2025-02-01
|
| Series: | ARO-The Scientific Journal of Koya University |
| Subjects: | |
| Online Access: | http://aro.koyauniversity.org/index.php/aro/article/view/1942 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850087560561295360 |
|---|---|
| author | Ismail Y. Maolood |
| author_facet | Ismail Y. Maolood |
| author_sort | Ismail Y. Maolood |
| collection | DOAJ |
| description |
Early diagnosis of cancer is crucial for improved patient results. With the aim of improving the effectiveness of cancer diagnosis, this paper introduces a new proposed method, computer-aided diagnosis, utilizing the level-set algorithm based on the edge detection approach for medical image segmentation. To assess the performance of our method, it was proven on a highly varied dataset that comprised liver cancer, Magnetic Resonance Imaging (MRI) brain cancer, and dermoscopy color images. By effectively integrating edge information into the level-set evolution process, the proposed method achieved impressive results. For liver cancer images, we obtained an accuracy of 0.9913, a sensitivity of 0.9165, and a Dice coefficient of 0.8820. Similarly, for dermoscopy color images, the method achieved an accuracy of 0.9979, a sensitivity of 0.9301, and a Dice coefficient of 0.9301. In the case of MRI images, the method demonstrated an accuracy of 0.9933, a sensitivity of 0.8591, and a Dice coefficient of 0.8591. The proposed method outperforms traditional techniques such as Simulated Annealing combined with Artificial Neural Network and Fuzzy Entropy with Level Set thresholding. This method demonstrates superior segmentation accuracy and robustness. By enabling precise identification of cancerous regions, this approach supports early diagnosis, reduces misdiagnosis, and enhances treatment planning, offering significant potential for improving cancer care and patient results.
|
| format | Article |
| id | doaj-art-3ba5cccdae864e50b8334b36d0e77cfa |
| institution | DOAJ |
| issn | 2410-9355 2307-549X |
| language | English |
| publishDate | 2025-02-01 |
| publisher | Koya University |
| record_format | Article |
| series | ARO-The Scientific Journal of Koya University |
| spelling | doaj-art-3ba5cccdae864e50b8334b36d0e77cfa2025-08-20T02:43:12ZengKoya UniversityARO-The Scientific Journal of Koya University2410-93552307-549X2025-02-0113110.14500/aro.11942Enhancing Cancer DiagnosisIsmail Y. Maolood0(1) Department of Computer Science, College of Science, Knowledge University, Erbil 44001, Kurdistan Region—F.R. Iraq; (2) Department of Information and Communication Technology Center, Ministry of Higher Education and Scientific Research, Erbil, Kurdistan Region—F.R. Iraq Early diagnosis of cancer is crucial for improved patient results. With the aim of improving the effectiveness of cancer diagnosis, this paper introduces a new proposed method, computer-aided diagnosis, utilizing the level-set algorithm based on the edge detection approach for medical image segmentation. To assess the performance of our method, it was proven on a highly varied dataset that comprised liver cancer, Magnetic Resonance Imaging (MRI) brain cancer, and dermoscopy color images. By effectively integrating edge information into the level-set evolution process, the proposed method achieved impressive results. For liver cancer images, we obtained an accuracy of 0.9913, a sensitivity of 0.9165, and a Dice coefficient of 0.8820. Similarly, for dermoscopy color images, the method achieved an accuracy of 0.9979, a sensitivity of 0.9301, and a Dice coefficient of 0.9301. In the case of MRI images, the method demonstrated an accuracy of 0.9933, a sensitivity of 0.8591, and a Dice coefficient of 0.8591. The proposed method outperforms traditional techniques such as Simulated Annealing combined with Artificial Neural Network and Fuzzy Entropy with Level Set thresholding. This method demonstrates superior segmentation accuracy and robustness. By enabling precise identification of cancerous regions, this approach supports early diagnosis, reduces misdiagnosis, and enhances treatment planning, offering significant potential for improving cancer care and patient results. http://aro.koyauniversity.org/index.php/aro/article/view/1942Cancer detectionComputer-aided diagnosisEdge detection techniquesLevel-set methodMedical image segmentation |
| spellingShingle | Ismail Y. Maolood Enhancing Cancer Diagnosis ARO-The Scientific Journal of Koya University Cancer detection Computer-aided diagnosis Edge detection techniques Level-set method Medical image segmentation |
| title | Enhancing Cancer Diagnosis |
| title_full | Enhancing Cancer Diagnosis |
| title_fullStr | Enhancing Cancer Diagnosis |
| title_full_unstemmed | Enhancing Cancer Diagnosis |
| title_short | Enhancing Cancer Diagnosis |
| title_sort | enhancing cancer diagnosis |
| topic | Cancer detection Computer-aided diagnosis Edge detection techniques Level-set method Medical image segmentation |
| url | http://aro.koyauniversity.org/index.php/aro/article/view/1942 |
| work_keys_str_mv | AT ismailymaolood enhancingcancerdiagnosis |