HLDnet: Novel deep learning based Artificial Intelligence tool fuses acetic acid and Lugol’s iodine cervicograms for accurate pre-cancer screening

CervicoXNet: an automated cervicogram interpretation network

Visual inspection with acetic acid (VIA) is a pre-cancerous screening program for low-middle-income countries (LMICs). Due to the limited number of oncology-gynecologist clinicians in LMICs, VIA examinations are performed mainly by medical workers. However, the inability of the medical workers to recognize a significant pattern based on cervicograms, VIA examination produces high inter-observer variance and high false-positive rate. This study proposed an automated cervicogram interpretation using explainable convolutional neural networks named "CervicoXNet" to support medical workers decision. The total number of 779 cervicograms was used for the learning process: 487 with VIA ( +) and 292 with VIA ( -). We performed data augmentation process under a geometric transformation scenario, such process produces 7325 cervicogram with VIA ( -) and 7242 cervicogram with VIA ( +). The proposed model outperformed other deep learning models, with 99.22% accuracy, 100% sensitivity, and 98.28% specificity. Moreover, to test the robustness of the proposed model, colposcope images used to validate the model's generalization ability. The results showed that the proposed architecture still produced satisfactory performance, with 98.11% accuracy, 98.33% sensitivity, and 98% specificity. It can be proven that the proposed model has been achieved satisfactory results. To make the prediction results visually interpretable, the results are localized with a heat map in fine-grained pixels using a combination of Grad-CAM and guided backpropagation. CervicoXNet can be used an alternative early screening tool with VIA alone.

HDFCN: A Robust Hybrid Deep Network Based on Feature Concatenation for Cervical Cancer Diagnosis on WSI Pap Smear Slides

Cervical cancer is a critical imperilment to a female's health due to its malignancy and fatality rate. The disease can be thoroughly cured by locating and treating the infected tissues in the preliminary phase. The traditional practice for screening cervical cancer is the examination of cervix tissues using the Papanicolaou (Pap) test. Manual inspection of pap smears involves false-negative outcomes due to human error even in the presence of the infected sample. Automated computer vision diagnosis revamps this obstacle and plays a substantial role in screening abnormal tissues affected due to cervical cancer. Here, in this paper, we propose a hybrid deep feature concatenated network (HDFCN) following two-step data augmentation to detect cervical cancer for binary and multiclass classification on the Pap smear images. This network carries out the classification of malignant samples for whole slide images (WSI) of the openly accessible SIPaKMeD database by utilizing the concatenation of features extracted from the fine-tuning of the deep learning (DL) models, namely, VGG-16, ResNet-152, and DenseNet-169, pretrained on the ImageNet dataset. The performance outcomes of the proposed model are compared with the individual performances of the aforementioned DL networks using transfer learning (TL). Our proposed model achieved an accuracy of 97.45% and 99.29% for 5-class and 2-class classifications, respectively. Additionally, the experiment is performed to classify liquid-based cytology (LBC) WSI data containing pap smear images.

Automated Precancerous Lesion Screening Using an Instance Segmentation Technique for Improving Accuracy

Precancerous screening using visual inspection with acetic acid (VIA) is suggested by the World Health Organization (WHO) for low-middle-income countries (LMICs). However, because of the limited number of gynecological oncologist clinicians in LMICs, VIA screening is primarily performed by general clinicians, nurses, or midwives (called medical workers). However, not being able to recognize the significant pathophysiology of human papilloma virus (HPV) infection in terms of the columnar epithelial-cell, squamous epithelial-cell, and white-spot regions with abnormal blood vessels may be further aggravated by VIA screening, which achieves a wide range of sensitivity (49-98%) and specificity (75-91%); this might lead to a false result and high interobserver variances. Hence, the automated detection of the columnar area (CA), subepithelial region of the squamocolumnar junction (SCJ), and acetowhite (AW) lesions is needed to support an accurate diagnosis. This study proposes a mask-RCNN architecture to simultaneously segment, classify, and detect CA and AW lesions. We conducted several experiments using 262 images of VIA+ cervicograms, and 222 images of VIA-cervicograms. The proposed model provided a satisfactory intersection over union performance for the CA of about 63.60%, and AW lesions of about 73.98%. The dice similarity coefficient performance was about 75.67% for the CA and about 80.49% for the AW lesion. It also performed well in cervical-cancer precursor-lesion detection, with a mean average precision of about 86.90% for the CA and of about 100% for the AW lesion, while also achieving 100% sensitivity and 92% specificity. Our proposed model with the instance segmentation approach can segment, detect, and classify cervical-cancer precursor lesions with satisfying performance only from a VIA cervicogram.