A deep learning approach for breast tissue malignancy diagnosis using micro-FTIR hyperspectral imaging

Página do item simplificado
dc.contributor.authorDEL-VALLE, MATHEUSpt_BR
dc.contributor.authorSANTOS, MOISES O. dospt_BR
dc.contributor.authorSANTOS, SOFIA N. dospt_BR
dc.contributor.authorBERNARDES, EMERSON S.pt_BR
dc.contributor.authorZEZELL, DENISE M.pt_BR
dc.coverageNacionalpt_BR
dc.creator.eventoENCONTRO DE OUTONO DA SOCIEDADE BRASILEIRA DE FÍSICA, 44.pt_BR
dc.date.accessioned2022-02-03T12:26:25Z
dc.date.available2022-02-03T12:26:25Z
dc.date.evento21-25 de junho, 2021pt_BR
dc.description.abstractThe breast cancer is the most incident cancer in women with an estimative of 2.1 million new cases in 2018. With the grown of deep learning techniques, several approaches in vibrational spectroscopy have been studied. In this way, this work aimed to classify breast samples as breast cancer or adenosis using a deep learning model. It was used the human breast cancer microarray BR804b (Biomax, Inc., USA), where one core of each group, cancer and adenosis, was imaged by a Cary Series 600 micro-FTIR imaging system (Agilent Technologies, USA). The system has a spatial resolution of 5.5 μm and about 100 thousand spectra were acquired for each group. The regions of interest were selected by two k-means clustering using amide I/II (1700 to 1500 cm-1) and highest paraffin intensity (1480 to 1450 cm-1) bands. Spectra were preprocessed by five steps: outlier removal using Hotelling’s T2 versus Q residuals; biofingerprint truncation; Savitzky–Golay filtering for smoothing and second derivative; Extended multiplicative signal correction (EMSC) with digital de-waxing; another outlier removal. The deep learning model was a convolutional neural network (CNN) fused with a fully connected neural network (FCNN). The CNN was built with 2 Conv1D-ReLU-MaxPooling1D-Dropout layers. The kernel size was set to 5 and dropout of 0.5. Dense layers were built by two layers of neurons-BatchNorm-ReLU-Dropout, with 100 and 50 neurons, dropout of 0.2. The output was a single neuron with sigmoid activation. Binary cross-entropy loss function was adopted with Adam optimizer. Accuracy metric was calculated during the training, where a threshold of 0.5 was applied on the output predictions. Model was trained by a 4-fold cross-validation by 20 epochs and using a batch size of 250. The train accuracy was 0.978/0.004 (mean/std), while the testing accuracy was 0.969/0.008, demonstrating a generalized model without overfitting. Accuracies near one indicate the proposed model as a potential technique for the breast cancer vs adenosis classification, where hyperparameters and the architecture should be optimized along higher sample number acquisition.pt_BR
dc.event.siglaEOSBFpt_BR
dc.identifier.citationDEL-VALLE, MATHEUS; SANTOS, MOISES O. dos; SANTOS, SOFIA N. dos; BERNARDES, EMERSON S.; ZEZELL, DENISE M. A deep learning approach for breast tissue malignancy diagnosis using micro-FTIR hyperspectral imaging. In: ENCONTRO DE OUTONO DA SOCIEDADE BRASILEIRA DE FÍSICA, 44., 21-25 de junho, 2021, Online. <b>Resumo...</b> São Paulo, SP: Sociedade Brasileira de Física, 2021. Disponível em: http://repositorio.ipen.br/handle/123456789/32690.
dc.identifier.orcid0000-0001-7404-9606pt_BR
dc.identifier.orcid0000-0002-0029-7313pt_BR
dc.identifier.orcidhttps://orcid.org/0000-0002-0029-7313
dc.identifier.orcidhttps://orcid.org/0000-0001-7404-9606
dc.identifier.urihttp://repositorio.ipen.br/handle/123456789/32690
dc.localSão Paulo, SPpt_BR
dc.local.eventoOnlinept_BR
dc.publisherSociedade Brasileira de Físicapt_BR
dc.rightsopenAccesspt_BR
dc.titleA deep learning approach for breast tissue malignancy diagnosis using micro-FTIR hyperspectral imagingpt_BR
dc.typeResumo de eventos científicospt_BR
dspace.entity.typePublication
ipen.autorSOFIA NASCIMENTO DOS SANTOS
ipen.autorEMERSON SOARES BERNARDES
ipen.autorDENISE MARIA ZEZELL
ipen.autorMOISES OLIVEIRA DOS SANTOS
ipen.autorMATHEUS DEL VALLE
ipen.codigoautor14464
ipen.codigoautor12099
ipen.codigoautor693
ipen.codigoautor8411
ipen.codigoautor15209
ipen.contributor.ipenauthorSOFIA NASCIMENTO DOS SANTOS
ipen.contributor.ipenauthorEMERSON SOARES BERNARDES
ipen.contributor.ipenauthorDENISE MARIA ZEZELL
ipen.contributor.ipenauthorMOISES OLIVEIRA DOS SANTOS
ipen.contributor.ipenauthorMATHEUS DEL VALLE
ipen.date.recebimento22-02
ipen.event.datapadronizada2021pt_BR
ipen.identifier.ipendoc28458pt_BR
ipen.notas.internasResumopt_BR
ipen.type.genreResumo
relation.isAuthorOfPublicationab78881a-78eb-42be-a463-aaf80e70de3d
relation.isAuthorOfPublication8115c8bd-822c-4f5a-9f49-3c12570ed40a
relation.isAuthorOfPublicationa565f8ad-3432-4891-98c0-a587f497db21
relation.isAuthorOfPublication1660cd3d-a7bb-40e2-9724-77f28d5c866a
relation.isAuthorOfPublicationfdd01116-8cc4-406a-aafb-606941dc28dc
relation.isAuthorOfPublication.latestForDiscoveryfdd01116-8cc4-406a-aafb-606941dc28dc
sigepi.autor.atividadeZEZELL, DENISE M.:693:920:Npt_BR
sigepi.autor.atividadeBERNARDES, EMERSON S.:12099:110:Npt_BR
sigepi.autor.atividadeSANTOS, SOFIA N. dos:14464:110:Npt_BR
sigepi.autor.atividadeSANTOS, MOISES O. dos:8411:920:Npt_BR
sigepi.autor.atividadeDEL-VALLE, MATHEUS:15209:920:Spt_BR
Pacote Original
Agora exibindo 1 - 1 de 1
Imagem de Miniatura
Nome:
28458.pdf
Tamanho:
173.83 KB
Formato:
Adobe Portable Document Format
Descrição:
Baixar
Licença do Pacote
Agora exibindo 1 - 1 de 1
Nenhuma Miniatura disponível
Nome:
license.txt
Tamanho:
1.71 KB
Formato:
Item-specific license agreed upon to submission
Descrição:
Baixar
Coleções
Eventos - Resumos