Multi-Modal Evolutionary Deep Learning Model for Ovarian Cancer Diagnosis

Abstract

Ovarian cancer (OC) is a common reason for mortality among women. Deep learning has recently proven better performance in predicting OC stages and subtypes. However, most of the state-of-the-art deep learning models employ single modality data, which may afford low-level performance due to insufficient representation of important OC characteristics. Furthermore, these deep learning models still lack to the optimization of the model construction, which requires high computational cost to train and deploy them. In this work, a hybrid evolutionary deep learning model, using multi-modal data, is proposed. The established multi-modal fusion framework amalgamates gene modality alongside with histopathological image modality. Based on the different states and forms of each modality, we set up deep feature extraction network, respectively. This includes a predictive antlion-optimized long-short-term-memory model to process gene longitudinal data. Another predictive antlion-optimized convolutional neural network model is included to process histopathology images. The topology of each customized feature network is automatically set by the antlion optimization algorithm to make it realize better performance. After that the output from the two improved networks is fused based upon weighted linear aggregation. The deep fused features are finally used to predict OC stage. A number of assessment indicators was used to compare the proposed model to other nine multi-modal fusion models constructed using distinct evolutionary algorithms. This was conducted using a benchmark for OC and two benchmarks for breast and lung cancers. The results reveal that the proposed model is more precise and accurate in diagnosing OC and the other cancers.

Computer vision and machine learning for medical image analysis: recent advances, challenges, and way forward

“…Common clinical uses of DL-based medical image classification methods include: skin cancer classification in dermoscopic images [56,57] , lung cancer identification in CT images [58] , breast cancer classification in mammograms [59] and ultrasound [60] images, brain cancer classification in MRI images [61,62] , diabetic retinopathy [63,64] , eye disease recognition in retinal fundus images [65] , and so on. In addition, the classification of histopathology images is also widely used for identifying different types of cancers such as colon cancer [66] , prostate cancer [67] , breast cancer [68] , and ovarian cancer [69] . Recently, DL-based methods have also been popularly utilised to identify COVID infections in the chest X-ray images [55] , and although the dataset for this application was limited to 150 patients, the deep neural network achieved 96% sensitivity.…”

“…Recent popular image classification architecture, InceptionResNetV2 has been used to identify retinal exudates and drusen in ultra-widefield fundus images [65] . A multiscale decision aggregation is used in [67] , pre-trained Inception-V3 in [68] , and a hybrid evolutionary DL in [69] to classify: prostate, breast, and ovarian cancer, respectively.…”

A Weakly Supervised Deep Learning Method for Guiding Ovarian Cancer Treatment and Identifying an Effective Biomarker

“…Wu et al [ 29 ] introduced automatic classification of ovarian cancer types from cytological images using deep convolutional neural networks. Ghoniem et al [ 30 ] built a multimodal evolutionary DL model for ovarian cancer diagnosis. Hong et al [ 31 ] built multiresolution deep learning models for predicting endometrial cancer subtypes and molecular features from histopathology images.…”

Systematic Analysis of Ovarian Cancer Empowered with Machine and Deep Learning: A Taxonomy and Future Challenges

“…Further k-means clustering used to cluster the subtype features so that gene could reduce for identifying the subtype of OC with logistic regression. A [65] deep learning multi-model framework implemented on histopathological and gene data to predict the stages of OC. This comprises two network models of feature extraction: Antlion optimizer-algorithm with long short-term memory-network and Antlion with convolutional neural network.…”