Supporting Radiologists with Automated Image Analysis: An Evaluation of Deep Learning Tools for Augmenting Breast Cancer Screening
Introduction: This study investigates the use of Artificial Neural Networks (ANNs) for analyzing breast cancer risk factors, aiming to improve predictive analytics in early diagnosis and prevention. By focusing on complex patterns among genetic, hormonal, lifestyle, and environmental factors, the...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SJORANM GmbH (Ltd.)
2025-04-01
|
| Series: | Swiss Journal of Radiology and Nuclear Medicine |
| Subjects: | |
| Online Access: | https://sjoranm.com/sjoranm/article/view/61 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850198707074498560 |
|---|---|
| author | Imene Bouharati Khaoula Bouharati |
| author_facet | Imene Bouharati Khaoula Bouharati |
| author_sort | Imene Bouharati |
| collection | DOAJ |
| description |
Introduction:
This study investigates the use of Artificial Neural Networks (ANNs) for analyzing breast cancer risk factors, aiming to improve predictive analytics in early diagnosis and prevention. By focusing on complex patterns among genetic, hormonal, lifestyle, and environmental factors, the objective is to determine how effectively ANNs can rank and assess these risks.
Methodology:
ANNs were applied to large datasets containing patient histories, medical records, and genetic information to evaluate their predictive power. The study leveraged deep learning techniques to process intricate, nonlinear relationships that traditional statistical approaches may overlook. Risk factors were analyzed to identify significant patterns, and the ANNs were tuned to optimize prediction accuracy and reliability.
Results and Discussion:
The results showed that ANNs could successfully identify key risk factors for breast cancer and rank them based on predictive strength. Deep learning techniques enhanced the accuracy of predictions, revealing subtle, nonlinear correlations among risk factors. However, challenges were noted in interpreting neural network models due to their complexity, and limitations in data quality and balance impacted outcomes. These findings highlight the advantages of ANNs in personalized risk assessment but emphasize the need for continued refinement to address interpretability issues.
Conclusion:
ANNs demonstrate considerable potential to advance breast cancer risk prediction, offering valuable insights for personalized prevention strategies. While further work is needed to optimize these models and integrate them effectively into clinical practice, ANNs could significantly enhance early risk assessment and improve outcomes in breast cancer.
|
| format | Article |
| id | doaj-art-e208efa22bfd43739562a9775e458b8c |
| institution | OA Journals |
| issn | 2813-7221 |
| language | English |
| publishDate | 2025-04-01 |
| publisher | SJORANM GmbH (Ltd.) |
| record_format | Article |
| series | Swiss Journal of Radiology and Nuclear Medicine |
| spelling | doaj-art-e208efa22bfd43739562a9775e458b8c2025-08-20T02:12:49ZengSJORANM GmbH (Ltd.)Swiss Journal of Radiology and Nuclear Medicine2813-72212025-04-0118110.59667/sjoranm.v18i1.16Supporting Radiologists with Automated Image Analysis: An Evaluation of Deep Learning Tools for Augmenting Breast Cancer ScreeningImene Bouharati0https://orcid.org/0009-0006-6412-4688Khaoula Bouharati1Faculty of Medicine, UFAS Setif 1 University, AlgeriaFaculty of Medicine, UFAS Setif 1 University, Algeria Introduction: This study investigates the use of Artificial Neural Networks (ANNs) for analyzing breast cancer risk factors, aiming to improve predictive analytics in early diagnosis and prevention. By focusing on complex patterns among genetic, hormonal, lifestyle, and environmental factors, the objective is to determine how effectively ANNs can rank and assess these risks. Methodology: ANNs were applied to large datasets containing patient histories, medical records, and genetic information to evaluate their predictive power. The study leveraged deep learning techniques to process intricate, nonlinear relationships that traditional statistical approaches may overlook. Risk factors were analyzed to identify significant patterns, and the ANNs were tuned to optimize prediction accuracy and reliability. Results and Discussion: The results showed that ANNs could successfully identify key risk factors for breast cancer and rank them based on predictive strength. Deep learning techniques enhanced the accuracy of predictions, revealing subtle, nonlinear correlations among risk factors. However, challenges were noted in interpreting neural network models due to their complexity, and limitations in data quality and balance impacted outcomes. These findings highlight the advantages of ANNs in personalized risk assessment but emphasize the need for continued refinement to address interpretability issues. Conclusion: ANNs demonstrate considerable potential to advance breast cancer risk prediction, offering valuable insights for personalized prevention strategies. While further work is needed to optimize these models and integrate them effectively into clinical practice, ANNs could significantly enhance early risk assessment and improve outcomes in breast cancer. https://sjoranm.com/sjoranm/article/view/61Artificial Neural Networks (ANNs)Predictive analyticsBreast cancerRisk factorsPersonalized prevention |
| spellingShingle | Imene Bouharati Khaoula Bouharati Supporting Radiologists with Automated Image Analysis: An Evaluation of Deep Learning Tools for Augmenting Breast Cancer Screening Swiss Journal of Radiology and Nuclear Medicine Artificial Neural Networks (ANNs) Predictive analytics Breast cancer Risk factors Personalized prevention |
| title | Supporting Radiologists with Automated Image Analysis: An Evaluation of Deep Learning Tools for Augmenting Breast Cancer Screening |
| title_full | Supporting Radiologists with Automated Image Analysis: An Evaluation of Deep Learning Tools for Augmenting Breast Cancer Screening |
| title_fullStr | Supporting Radiologists with Automated Image Analysis: An Evaluation of Deep Learning Tools for Augmenting Breast Cancer Screening |
| title_full_unstemmed | Supporting Radiologists with Automated Image Analysis: An Evaluation of Deep Learning Tools for Augmenting Breast Cancer Screening |
| title_short | Supporting Radiologists with Automated Image Analysis: An Evaluation of Deep Learning Tools for Augmenting Breast Cancer Screening |
| title_sort | supporting radiologists with automated image analysis an evaluation of deep learning tools for augmenting breast cancer screening |
| topic | Artificial Neural Networks (ANNs) Predictive analytics Breast cancer Risk factors Personalized prevention |
| url | https://sjoranm.com/sjoranm/article/view/61 |
| work_keys_str_mv | AT imenebouharati supportingradiologistswithautomatedimageanalysisanevaluationofdeeplearningtoolsforaugmentingbreastcancerscreening AT khaoulabouharati supportingradiologistswithautomatedimageanalysisanevaluationofdeeplearningtoolsforaugmentingbreastcancerscreening |