A Study on the Effectiveness of Deep Learning-Based Anomaly Detection Methods for Breast Ultrasonography
In the medical field, it is delicate to anticipate good performance in using deep learning due to the lack of large-scale training data and class imbalance. In particular, ultrasound, which is a key breast cancer diagnosis method, is delicate to diagnose accurately as the quality and interpretation...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-03-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/23/5/2864 |
| _version_ | 1827752033700020224 |
|---|---|
| author | Changhee Yun Bomi Eom Sungjun Park Chanho Kim Dohwan Kim Farah Jabeen Won Hwa Kim Hye Jung Kim Jaeil Kim |
| author_facet | Changhee Yun Bomi Eom Sungjun Park Chanho Kim Dohwan Kim Farah Jabeen Won Hwa Kim Hye Jung Kim Jaeil Kim |
| author_sort | Changhee Yun |
| collection | DOAJ |
| description | In the medical field, it is delicate to anticipate good performance in using deep learning due to the lack of large-scale training data and class imbalance. In particular, ultrasound, which is a key breast cancer diagnosis method, is delicate to diagnose accurately as the quality and interpretation of images can vary depending on the operator’s experience and proficiency. Therefore, computer-aided diagnosis technology can facilitate diagnosis by visualizing abnormal information such as tumors and masses in ultrasound images. In this study, we implemented deep learning-based anomaly detection methods for breast ultrasound images and validated their effectiveness in detecting abnormal regions. Herein, we specifically compared the sliced-Wasserstein autoencoder with two representative unsupervised learning models autoencoder and variational autoencoder. The anomalous region detection performance is estimated with the normal region labels. Our experimental results showed that the sliced-Wasserstein autoencoder model outperformed the anomaly detection performance of others. However, anomaly detection using the reconstruction-based approach may not be effective because of the occurrence of numerous false-positive values. In the following studies, reducing these false positives becomes an important challenge. |
| first_indexed | 2024-03-11T07:09:05Z |
| format | Article |
| id | doaj.art-0036d7f1f6db4ce789f4a1a15516c846 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-11T07:09:05Z |
| publishDate | 2023-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-0036d7f1f6db4ce789f4a1a15516c8462023-11-17T08:40:53ZengMDPI AGSensors1424-82202023-03-01235286410.3390/s23052864A Study on the Effectiveness of Deep Learning-Based Anomaly Detection Methods for Breast UltrasonographyChanghee Yun0Bomi Eom1Sungjun Park2Chanho Kim3Dohwan Kim4Farah Jabeen5Won Hwa Kim6Hye Jung Kim7Jaeil Kim8National Information Society Agency, Daegu 41068, Republic of KoreaNational Information Society Agency, Daegu 41068, Republic of KoreaSchool of Computer Science and Engineering, Kyungpook National University, Daegu 41566, Republic of KoreaSchool of Computer Science and Engineering, Kyungpook National University, Daegu 41566, Republic of KoreaDepartment of Artificial Intelligence, Kyungpook National University, Daegu 41566, Republic of KoreaSchool of Computer Science and Engineering, Kyungpook National University, Daegu 41566, Republic of KoreaDepartment of Radiology, Kyungpook National University Chilgok Hospital, Kyungpook National University, Daegu 41404, Republic of KoreaDepartment of Radiology, Kyungpook National University Chilgok Hospital, Kyungpook National University, Daegu 41404, Republic of KoreaSchool of Computer Science and Engineering, Kyungpook National University, Daegu 41566, Republic of KoreaIn the medical field, it is delicate to anticipate good performance in using deep learning due to the lack of large-scale training data and class imbalance. In particular, ultrasound, which is a key breast cancer diagnosis method, is delicate to diagnose accurately as the quality and interpretation of images can vary depending on the operator’s experience and proficiency. Therefore, computer-aided diagnosis technology can facilitate diagnosis by visualizing abnormal information such as tumors and masses in ultrasound images. In this study, we implemented deep learning-based anomaly detection methods for breast ultrasound images and validated their effectiveness in detecting abnormal regions. Herein, we specifically compared the sliced-Wasserstein autoencoder with two representative unsupervised learning models autoencoder and variational autoencoder. The anomalous region detection performance is estimated with the normal region labels. Our experimental results showed that the sliced-Wasserstein autoencoder model outperformed the anomaly detection performance of others. However, anomaly detection using the reconstruction-based approach may not be effective because of the occurrence of numerous false-positive values. In the following studies, reducing these false positives becomes an important challenge.https://www.mdpi.com/1424-8220/23/5/2864breast cancerultrasonographydeep learninganomaly detectionautoencoder |
| spellingShingle | Changhee Yun Bomi Eom Sungjun Park Chanho Kim Dohwan Kim Farah Jabeen Won Hwa Kim Hye Jung Kim Jaeil Kim A Study on the Effectiveness of Deep Learning-Based Anomaly Detection Methods for Breast Ultrasonography Sensors breast cancer ultrasonography deep learning anomaly detection autoencoder |
| title | A Study on the Effectiveness of Deep Learning-Based Anomaly Detection Methods for Breast Ultrasonography |
| title_full | A Study on the Effectiveness of Deep Learning-Based Anomaly Detection Methods for Breast Ultrasonography |
| title_fullStr | A Study on the Effectiveness of Deep Learning-Based Anomaly Detection Methods for Breast Ultrasonography |
| title_full_unstemmed | A Study on the Effectiveness of Deep Learning-Based Anomaly Detection Methods for Breast Ultrasonography |
| title_short | A Study on the Effectiveness of Deep Learning-Based Anomaly Detection Methods for Breast Ultrasonography |
| title_sort | study on the effectiveness of deep learning based anomaly detection methods for breast ultrasonography |
| topic | breast cancer ultrasonography deep learning anomaly detection autoencoder |
| url | https://www.mdpi.com/1424-8220/23/5/2864 |
| work_keys_str_mv | AT changheeyun astudyontheeffectivenessofdeeplearningbasedanomalydetectionmethodsforbreastultrasonography AT bomieom astudyontheeffectivenessofdeeplearningbasedanomalydetectionmethodsforbreastultrasonography AT sungjunpark astudyontheeffectivenessofdeeplearningbasedanomalydetectionmethodsforbreastultrasonography AT chanhokim astudyontheeffectivenessofdeeplearningbasedanomalydetectionmethodsforbreastultrasonography AT dohwankim astudyontheeffectivenessofdeeplearningbasedanomalydetectionmethodsforbreastultrasonography AT farahjabeen astudyontheeffectivenessofdeeplearningbasedanomalydetectionmethodsforbreastultrasonography AT wonhwakim astudyontheeffectivenessofdeeplearningbasedanomalydetectionmethodsforbreastultrasonography AT hyejungkim astudyontheeffectivenessofdeeplearningbasedanomalydetectionmethodsforbreastultrasonography AT jaeilkim astudyontheeffectivenessofdeeplearningbasedanomalydetectionmethodsforbreastultrasonography AT changheeyun studyontheeffectivenessofdeeplearningbasedanomalydetectionmethodsforbreastultrasonography AT bomieom studyontheeffectivenessofdeeplearningbasedanomalydetectionmethodsforbreastultrasonography AT sungjunpark studyontheeffectivenessofdeeplearningbasedanomalydetectionmethodsforbreastultrasonography AT chanhokim studyontheeffectivenessofdeeplearningbasedanomalydetectionmethodsforbreastultrasonography AT dohwankim studyontheeffectivenessofdeeplearningbasedanomalydetectionmethodsforbreastultrasonography AT farahjabeen studyontheeffectivenessofdeeplearningbasedanomalydetectionmethodsforbreastultrasonography AT wonhwakim studyontheeffectivenessofdeeplearningbasedanomalydetectionmethodsforbreastultrasonography AT hyejungkim studyontheeffectivenessofdeeplearningbasedanomalydetectionmethodsforbreastultrasonography AT jaeilkim studyontheeffectivenessofdeeplearningbasedanomalydetectionmethodsforbreastultrasonography |