Multimodal Melanoma Detection with Federated Learning

Agbley, Bless Lord Y.; Li, Minglu; Haq, Amin Ul; Bankas, Edem Kwedzo; Ahmad, Sultan; Agyemang, Isaac Osei; Kulevome, Delanyo Kwame Bensah; Ndiaye, Waldiodio David; Cobbinah, Bernard M.; Latipova, Shoistamo

doi:10.1109/iccwamtip53232.2021.9674116

Cited by 25 publications

(27 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The various Otsu methods discussed as thresholding-based improvised histogram, K-means, etc., along with their advantages and disadvantages. This method is mostly used to reduce the complexity of 1-D and 2-D. Agbley et al [ 12 ] and Singh and Veenadhari [ 13 ] gave hybrid technology for segmenting out ROI by merging the region and global thresholding applied to the mammographic images. To eliminate Gaussian noise, Wiener filters were used, and then the resulting image was normalized using the histogram to enhance the quality of input images.…”

Section: Literature Reviewmentioning

confidence: 99%

Ensemble Learning-Based Hybrid Segmentation of Mammographic Images for Breast Cancer Risk Prediction Using Fuzzy C-Means and CNN Model

Jha

Ahmad

Arya

et al. 2023

Journal of Healthcare Engineering

Self Cite

View full text Add to dashboard Cite

The research interest in this field is that females are not aware of their health conditions until they develop tumour, especially when breast cancer is concerned. The breast cancer risk factors include genetics, heredity, and sedentary lifestyle. The prime concern for the mortality rate among females is breast cancer, and breast cancer is on the rise, both in rural and urban India. Women aged 45 or above are more vulnerable to this disease. Images are more effective at depicting information as compared to text. With the advancement in technology, several computerized techniques have come up to extract hidden information from the images. The processed images have found their application in several sectors and medical science is one of them. Disease-like breast cancer affects most women universally and it happens due to the existence of breast masses in the breast region for the development of breast cancer in women. Timely breast cancer detection can also increase the rate of effective treatment and the survival of women suffering from breast cancer. This work elaborates the method of performing hybrid segmentation techniques using CLAHE, morphological operations on mammogram images, and classified images using deep learning. Images from the MIAS database have been used to obtain readings for parameters: threshold, accuracy, sensitivity, specificity rate, biopsy rate, or a combination of all the parameters and many others under study.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Ensemble Learning-Based Hybrid Segmentation of Mammographic Images for Breast Cancer Risk Prediction Using Fuzzy C-Means and CNN Model

Jha

Ahmad

Arya

et al. 2023

Journal of Healthcare Engineering

Self Cite

View full text Add to dashboard Cite

show abstract

“…Along with this, the investigation on the applicability of federated learning to different fields [16][17][18][19] has become an interesting research area. To note some of those, in the medical domain, a multi-modal approach to detect Covid-19 using the combination of information from X-ray and Ultrasound images is demonstrated in.…”

Section: Related Literaturementioning

confidence: 99%

Towards developing a data security aware federated training framework in multi-modal contested environments

Ovi

Dey

Roy

et al. 2022

Artificial Intelligence and Machine Learning for Multi-Domain Operations Applications IV

View full text Add to dashboard Cite

“…Institutions are usually obliged to run models on their localized datasets to keep their patients’ data private. Motivated by the successful use of FL by Google to achieve a high performing model for predicting words on their Gboard [ 31 , 32 ], some recent works in computerized medical image analysis [ 19 ] have adopted FL to help protect patients’ privacy. This work also uses FL to bridge the IC-NST data availability gap and to encourage collaboration between different institutions without sharing their patients’ data.…”

Section: Related Workmentioning

confidence: 99%

“…Classifier: The features obtained from the ResNet

and the features extracted with the GaborNet

are fused [ 19 , 42 ] and fed into a classifier for prediction. The classifier consists of a linear layer with a 256 output size, a ReLU activation layer, a batch normalization layer, a dropout of 0.5, and a final linear layer with an output of 2 for the two classes.…”

Section: Proposed Approachmentioning

confidence: 99%

“…Capturing these specimens into images makes them available for use in computer-aided detection/diagnosis [ 14 ]. Approaches such as machine learning (ML) and, more specifically, deep learning (DL), have been explored in recent years due to their successes in aiding in the prognosis and diagnosis of other medical conditions [ 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 ]. These techniques provide mathematical models for automating the detection process.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Federated Learning-Based Detection of Invasive Carcinoma of No Special Type with Histopathological Images

Agbley

Hossin

et al. 2022

Diagnostics

Self Cite

View full text Add to dashboard Cite

Invasive carcinoma of no special type (IC-NST) is known to be one of the most prevalent kinds of breast cancer, hence the growing research interest in studying automated systems that can detect the presence of breast tumors and appropriately classify them into subtypes. Machine learning (ML) and, more specifically, deep learning (DL) techniques have been used to approach this problem. However, such techniques usually require massive amounts of data to obtain competitive results. This requirement makes their application in specific areas such as health problematic as privacy concerns regarding the release of patients’ data publicly result in a limited number of publicly available datasets for the research community. This paper proposes an approach that leverages federated learning (FL) to securely train mathematical models over multiple clients with local IC-NST images partitioned from the breast histopathology image (BHI) dataset to obtain a global model. First, we used residual neural networks for automatic feature extraction. Then, we proposed a second network consisting of Gabor kernels to extract another set of features from the IC-NST dataset. After that, we performed a late fusion of the two sets of features and passed the output through a custom classifier. Experiments were conducted for the federated learning (FL) and centralized learning (CL) scenarios, and the results were compared. Competitive results were obtained, indicating the positive prospects of adopting FL for IC-NST detection. Additionally, fusing the Gabor features with the residual neural network features resulted in the best performance in terms of accuracy, F1 score, and area under the receiver operation curve (AUC-ROC). The models show good generalization by performing well on another domain dataset, the breast cancer histopathological (BreakHis) image dataset. Our method also outperformed other methods from the literature.

show abstract

Multimodal Melanoma Detection with Federated Learning

Cited by 25 publications

References 20 publications

Ensemble Learning-Based Hybrid Segmentation of Mammographic Images for Breast Cancer Risk Prediction Using Fuzzy C-Means and CNN Model

Ensemble Learning-Based Hybrid Segmentation of Mammographic Images for Breast Cancer Risk Prediction Using Fuzzy C-Means and CNN Model

Towards developing a data security aware federated training framework in multi-modal contested environments

Federated Learning-Based Detection of Invasive Carcinoma of No Special Type with Histopathological Images

Contact Info

Product

Resources

About