Ensemble of Convolutional Neural Networks for Classification of Breast Microcalcification from Mammograms

Date

2017-07-15

Author

SERT, Egemen
Ertekin Bolelli, Şeyda
Halıcı, Uğur

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

137
views

0
downloads

Human level recall performance in detecting breast cancer considering microcalcifications from mammograms has a recall value between 74.5% and 92.3%. In this research, we approach to breast microcalcification classification problem using convolutional neural networks along with various preprocessing methods such as contrast scaling, dilation, cropping etc. and decision fusion using ensemble of networks. Various experiments on Digital Database for Screening Mammography dataset showed that preprocessing poses great importance on the classification performance. The stand-alone models using the dilation and cropping preprocessing techniques achieved the highest recall value of 91.3%. The ensembles of the stand-alone models surpass this recall value and a 97.3% value of recall is achieved. The ensemble having the highest F1 Score (harmonic mean of precision and recall), which is 94.5%, has a recall value of 94.0% and a precision value of 95.0%. This recall is still above human level performance and the models achieve competitive results in terms of accuracy, precision, recall and F1 score measures.

Subject Keywords

Mammograms, Microcalcification, Deep Learning, Computer Vision, Convolutional Neural Networks, Decision Fusion, Ensemble Of Networks

URI

https://hdl.handle.net/11511/47785

DOI

https://doi.org/10.1109/embc.2017.8036918

Collections

Department of Computer Engineering, Conference / Seminar

Suggestions

OpenMETU
Core

Using Deep Learning for Mammography Classification Hepsag, Pinar Uskaner; ÖZEL, SELMA AYŞE; Yazıcı, Adnan (2017-10-08) Breast biopsies based on the results of mammography and ultrasound have been diagnosed as benign at a rate of approximately 40 to 60 percent. Negative biopsy results have negative impacts on many aspects such as unnecessary operations, fear, pain, and cost. Therefore, there is a need for a more reliable technique to reduce the number of unnecessary biopsies in the diagnosis of breast cancer. So, computer-aided diagnostic methods are very important for doctors to make more accurate decisions and to avoid unn...
Examination of the dielectrophoretic spectra of MCF7 breast cancer cells and leukocytes Çağlayan, Zeynep; Demircan Yalçın, Yağmur; Külah, Haluk (Wiley, 2020-03-01) The detection of circulating tumor cells (CTCs) in blood is crucial to assess metastatic progression and to guide therapy. Dielectrophoresis (DEP) is a powerful cell surface marker-free method that allows intrinsic dielectric properties of suspended cells to be exploited for CTC enrichment/isolation from blood. Design of a successful DEP-based CTC enrichment/isolation system requires that the DEP response of the targeted particles should accurately be known. This paper presents a DEP spectrum method to inve...
The serum immunoglobulin G glycosylation signature of gastric cancer Ruhaak, L. Renee; Barkauskas, Donald A.; Torres, Javier; Cooke, Cara L.; Wu, Lauren D.; Stroble, Carol; Özcan Kabasakal, Süreyya; Williams, Cynthia C.; Camorlinga, Margarita; Rocke, David M.; Lebrilla, Carlito B.; Solnick, Jay V. (Elsevier BV, 2015-03-01) Biomarkers may facilitate detection of gastric cancer at an earlier stage and reduce mortality. Here we sought to determine if the glycosylation profile of serum immunoglobulin G (IgG) could distinguish patients with non-atrophic gastritis (NAG), duodenal ulcer (DU) and gastric cancer (GC). Serum IgG was released and analyzed using nano-LC–TOF mass spectrometry. Statistically significant false discovery rate (FDR)-adjusted p-values were observed for 18 glycans, eight that differed significantly between NAG ...
Effect of input size on the classification of lung nodules using convolutional neural networks Akciǧer nodüllerinin evrişimsel sinir aǧlari kullanilarak siniflandirilmasinda girdi boyutunun etkisi POLAT, Gorkem; Serinağaoğlu Doğrusöz, Yeşim; Halıcı, Uğur (2018-05-05) Recent studies have shown that lung cancer screening using annual low-dose computed tomography (CT) reduces lung cancer mortality by 20% compared to traditional chest radiography. Therefore, CT lung screening has started to be used widely all across the world. However, analyzing these images is a serious burden for radiologists. The number of slices in a CT scan can be up to 600. Therefore, computer-aided-detection (CAD) systems are very important for faster and more accurate assessment of the data. In this...
Classification of lung nodules in CT images using convolutional neural networks Polat, Görkem; Serinağaoğlu Doğrusöz, Yeşim; Halıcı, Uğur; Department of Electrical and Electronics Engineering (2018) Recent studies have shown that lung cancer screening using annual low-dose computed tomography (CT) reduces lung cancer mortality by 20% compared to traditional chest radiography. Therefore, CT lung screening has started to be used widely all across the world. However, analyzing these images is a serious burden for radiologists. The number of slices in a CT scan can be up to 600. Therefore, computeraided-detection (CAD) systems are very important for faster and more accurate assessment of the data. In this ...

Citation Formats

E. SERT, Ş. Ertekin Bolelli, and U. Halıcı, “Ensemble of Convolutional Neural Networks for Classification of Breast Microcalcification from Mammograms,” 2017, Accessed: 00, 2020. [Online]. Available: https://hdl.handle.net/11511/47785.