An Observational Study of Deep Learning and Automated Evaluation of Cervical Images for Cancer Screening

Deep Learning Diagnostic Classification of Cervical Images to Augment Colposcopic Impression

OBJECTIVE

A deep learning classifier that improves the accuracy of colposcopic impression.

METHODS

Colposcopy images taken 56 seconds after acetic acid application were processed by a cervix detection algorithm to identify the cervical region. We optimized models based on the SegFormer architecture to classify each cervix as high-grade or negative/low-grade. The data were split into histologically stratified, random training, validation, and test subsets (80%-10%-10%). We replicated a 10-fold experiment to align with a prior study utilizing expert reviewer analysis of the same images. To evaluate the model's robustness across different cameras, we retrained it after dividing the dataset by camera type. Subsequently, we retrained the model on a new, histologically stratified random data split and integrated the results with patients' age and referral data to train a Gradient Boosted Tree model for final classification. Model accuracy was assessed by the receiver operating characteristic area under the curve (AUC), Youden's index (YI), sensitivity, and specificity compared to the histology.

RESULTS

Out of 5,485 colposcopy images, 4,946 with histology and a visible cervix were used. The model's average performance in the 10-fold experiment was AUC = 0.75, YI = 0.37 (sensitivity = 63%, specificity = 74%), outperforming the experts' average YI of 0.16. Transferability across camera types was effective, with AUC = 0.70, YI = 0.33. Integrating image-based predictions with referral data improved outcomes to AUC = 0.81 and YI = 0.46. The use of model predictions alongside the original colposcopic impression boosted overall performance.

CONCLUSIONS

Deep learning cervical image classification demonstrated robustness and outperformed experts. Further improved by including additional patient information, it shows potential for clinical utility complementing colposcopy.

Clinical evaluation of an artificial intelligence-assisted cytological system among screening strategies for a cervical cancer high-risk population

BACKGROUND

Primary cervical cancer screening and treating precancerous lesions are effective ways to prevent cervical cancer. However, the coverage rates of human papillomavirus (HPV) vaccines and routine screening are low in most developing countries and even some developed countries. This study aimed to explore the benefit of an artificial intelligence-assisted cytology (AI) system in a screening program for a cervical cancer high-risk population in China.

METHODS

A total of 1231 liquid-based cytology (LBC) slides from women who underwent colposcopy at the Chinese PLA General Hospital from 2018 to 2020 were collected. All women had received a histological diagnosis based on the results of colposcopy and biopsy. The sensitivity (Se), specificity (Sp), positive predictive value (PPV), negative predictive value (NPV), false-positive rate (FPR), false-negative rate (FNR), overall accuracy (OA), positive likelihood ratio (PLR), negative likelihood ratio (NLR) and Youden index (YI) of the AI, LBC, HPV, LBC + HPV, AI + LBC, AI + HPV and HPV Seq LBC screening strategies at low-grade squamous intraepithelial lesion (LSIL) and high-grade squamous intraepithelial lesion (HSIL) thresholds were calculated to assess their effectiveness. Receiver operating characteristic (ROC) curve analysis was conducted to assess the diagnostic values of the different screening strategies.

RESULTS

The Se and Sp of the primary AI-alone strategy at the LSIL and HSIL thresholds were superior to those of the LBC + HPV cotesting strategy. Among the screening strategies, the YIs of the AI strategy at the LSIL + threshold and HSIL + threshold were the highest. At the HSIL + threshold, the AI strategy achieved the best result, with an AUC value of 0.621 (95% CI, 0.587-0.654), whereas HPV testing achieved the worst result, with an AUC value of 0.521 (95% CI, 0.484-0.559). Similarly, at the LSIL + threshold, the LBC-based strategy achieved the best result, with an AUC of 0.637 (95% CI, 0.606-0.668), whereas HPV testing achieved the worst result, with an AUC of 0.524 (95% CI, 0.491-0.557). Moreover, the AUCs of the AI and LBC strategies at this threshold were similar (0.631 and 0.637, respectively).

CONCLUSIONS

These results confirmed that AI-only screening was the most authoritative method for diagnosing HSILs and LSILs, improving the accuracy of colposcopy diagnosis, and was more beneficial for patients than traditional LBC + HPV cotesting.

Artificial intelligence enables precision diagnosis of cervical cytology grades and cervical cancer

Cervical cancer is a significant global health issue, its prevalence and prognosis highlighting the importance of early screening for effective prevention. This research aimed to create and validate an artificial intelligence cervical cancer screening (AICCS) system for grading cervical cytology. The AICCS system was trained and validated using various datasets, including retrospective, prospective, and randomized observational trial data, involving a total of 16,056 participants. It utilized two artificial intelligence (AI) models: one for detecting cells at the patch-level and another for classifying whole-slide image (WSIs). The AICCS consistently showed high accuracy in predicting cytology grades across different datasets. In the prospective assessment, it achieved an area under curve (AUC) of 0.947, a sensitivity of 0.946, a specificity of 0.890, and an accuracy of 0.892. Remarkably, the randomized observational trial revealed that the AICCS-assisted cytopathologists had a significantly higher AUC, specificity, and accuracy than cytopathologists alone, with a notable 13.3% enhancement in sensitivity. Thus, AICCS holds promise as an additional tool for accurate and efficient cervical cancer screening.

Application of convolutional neural network for differentiating ovarian thecoma-fibroma and solid ovarian cancer based on MRI

BACKGROUND

Ovarian thecoma-fibroma and solid ovarian cancer have similar clinical and imaging features, and it is difficult for radiologists to differentiate them. Since the treatment and prognosis of them are different, accurate characterization is crucial.

PURPOSE

To non-invasively differentiate ovarian thecoma-fibroma and solid ovarian cancer by convolutional neural network based on magnetic resonance imaging (MRI), and to provide the interpretability of the model.

MATERIAL AND METHODS

A total of 156 tumors, including 86 ovarian thecoma-fibroma and 70 solid ovarian cancer, were split into the training set, the validation set, and the test set according to the ratio of 8:1:1 by stratified random sampling. In this study, we used four different networks, two different weight modes, two different optimizers, and four different sizes of regions of interest (ROI) to test the model performance. This process was repeated 10 times to calculate the average performance of the test set. The gradient weighted class activation mapping (Grad-CAM) was used to explain how the model makes classification decisions by visual location map.

RESULTS

ResNet18, which had pre-trained weight, using Adam and one multiple ROI circumscribed rectangle, achieved best performance. The average accuracy, precision, recall, and AUC were 0.852, 0.828, 0.848, and 0.919 (P < 0.01), respectively. Grad-CAM showed areas associated with classification appeared on the edge or interior of ovarian thecoma-fibroma and the interior of solid ovarian cancer.

CONCLUSION

This study shows that convolution neural network based on MRI can be helpful for radiologists in differentiating ovarian thecoma-fibroma and solid ovarian cancer.

Lightweight Low-Rank Adaptation Vision Transformer Framework for Cervical Cancer Detection and Cervix Type Classification

Cervical cancer is a major health concern worldwide, highlighting the urgent need for better early detection methods to improve outcomes for patients. In this study, we present a novel digital pathology classification approach that combines Low-Rank Adaptation (LoRA) with the Vision Transformer (ViT) model. This method is aimed at making cervix type classification more efficient through a deep learning classifier that does not require as much data. The key innovation is the use of LoRA, which allows for the effective training of the model with smaller datasets, making the most of the ability of ViT to represent visual information. This approach performs better than traditional Convolutional Neural Network (CNN) models, including Residual Networks (ResNets), especially when it comes to performance and the ability to generalize in situations where data are limited. Through thorough experiments and analysis on various dataset sizes, we found that our more streamlined classifier is highly accurate in spotting various cervical anomalies across several cases. This work advances the development of sophisticated computer-aided diagnostic systems, facilitating more rapid and accurate detection of cervical cancer, thereby significantly enhancing patient care outcomes.

Artificial intelligence for cervical cancer screening: Scoping review, 2009-2022

BACKGROUND

The intersection of artificial intelligence (AI) with cancer research is increasing, and many of the advances have focused on the analysis of cancer images.

OBJECTIVES

To describe and synthesize the literature on the diagnostic accuracy of AI in early imaging diagnosis of cervical cancer following Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews (PRISMA-ScR).

SEARCH STRATEGY

Arksey and O'Malley methodology was used and PubMed, Scopus, and Google Scholar databases were searched using a combination of English and Spanish keywords.

SELECTION CRITERIA

Identified titles and abstracts were screened to select original reports and cross-checked for overlap of cases.

DATA COLLECTION AND ANALYSIS

A descriptive summary was organized by the AI algorithm used, total of images analyzed, data source, clinical comparison criteria, and diagnosis performance.

MAIN RESULTS

We identified 32 studies published between 2009 and 2022. The primary sources of images were digital colposcopy, cervicography, and mobile devices. The machine learning/deep learning (DL) algorithms applied in the articles included support vector machine (SVM), random forest classifier, k-nearest neighbors, multilayer perceptron, C4.5, Naïve Bayes, AdaBoost, XGboots, conditional random fields, Bayes classifier, convolutional neural network (CNN; and variations), ResNet (several versions), YOLO+EfficientNetB0, and visual geometry group (VGG; several versions). SVM and DL methods (CNN, ResNet, VGG) showed the best diagnostic performances, with an accuracy of over 97%.

CONCLUSION

We concluded that the use of AI for cervical cancer screening has increased over the years, and some results (mainly from DL) are very promising. However, further research is necessary to validate these findings.

Cost-effectiveness of single-visit cervical cancer screening in KwaZulu-Natal, South Africa: a model-based analysis accounting for the HIV epidemic

Introduction

Women living with human immunodeficiency virus (WLHIV) face elevated risks of human papillomavirus (HPV) acquisition and cervical cancer (CC). Coverage of CC screening and treatment remains low in low-and-middle-income settings, reflecting resource challenges and loss to follow-up with current strategies. We estimated the health and economic impact of alternative scalable CC screening strategies in KwaZulu-Natal, South Africa, a region with high burden of CC and HIV.

Methods

We parameterized a dynamic compartmental model of HPV and HIV transmission and CC natural history to KwaZulu-Natal. Over 100 years, we simulated the status quo of a multi-visit screening and treatment strategy with cytology and colposcopy triage (South African standard of care) and six single-visit comparator scenarios with varying: 1) screening strategy (HPV DNA testing alone, with genotyping, or with automated visual evaluation triage, a new high-performance technology), 2) screening frequency (once-per-lifetime for all women, or repeated every 5 years for WLHIV and twice for women without HIV), and 3) loss to follow-up for treatment. Using the Ministry of Health perspective, we estimated costs associated with HPV vaccination, screening, and pre-cancer, CC, and HIV treatment. We quantified CC cases, deaths, and disability-adjusted life-years (DALYs) averted for each scenario. We discounted costs (2022 US dollars) and outcomes at 3% annually and calculated incremental cost-effectiveness ratios (ICERs).

Results

We projected 69,294 new CC cases and 43,950 CC-related deaths in the status quo scenario. HPV DNA testing achieved the greatest improvement in health outcomes, averting 9.4% of cases and 9.0% of deaths with one-time screening and 37.1% and 35.1%, respectively, with repeat screening. Compared to the cost of the status quo ($12.79 billion), repeat screening using HPV DNA genotyping had the greatest increase in costs. Repeat screening with HPV DNA testing was the most effective strategy below the willingness to pay threshold (ICER: $3,194/DALY averted). One-time screening with HPV DNA testing was also an efficient strategy (ICER: $1,398/DALY averted).

Conclusions

Repeat single-visit screening with HPV DNA testing was the optimal strategy simulated. Single-visit strategies with increased frequency for WLHIV may be cost-effective in KwaZulu-Natal and similar settings with high HIV and HPV prevalence.

Serum Proteomic Signatures in Cervical Cancer: Current Status and Future Directions

In 2020, the World Health Organization (WHO) reported 604,000 new diagnoses of cervical cancer (CC) worldwide, and over 300,000 CC-related fatalities. The vast majority of CC cases are caused by persistent human papillomavirus (HPV) infections. HPV-related CC incidence and mortality rates have declined worldwide because of increased HPV vaccination and CC screening with the Papanicolaou test (PAP test). Despite these significant improvements, developing countries face difficulty implementing these programs, while developed nations are challenged with identifying HPV-independent cases. Molecular and proteomic information obtained from blood or tumor samples have a strong potential to provide information on malignancy progression and response to therapy in CC. There is a large amount of published biomarker data related to CC available but the extensive validation required by the FDA approval for clinical use is lacking. The ability of researchers to use the big data obtained from clinical studies and to draw meaningful relationships from these data are two obstacles that must be overcome for implementation into clinical practice. We report on identified multimarker panels of serum proteomic studies in CC for the past 5 years, the potential for modern computational biology efforts, and the utilization of nationwide biobanks to bridge the gap between multivariate protein signature development and the prediction of clinically relevant CC patient outcomes.

Development and validation of artificial intelligence-based analysis software to support screening system of cervical intraepithelial neoplasia

Cervical cancer, the fourth most common cancer among women worldwide, often proves fatal and stems from precursor lesions caused by high-risk human papillomavirus (HR-HPV) infection. Accurate and early diagnosis is crucial for effective treatment. Current screening methods, such as the Pap test, liquid-based cytology (LBC), visual inspection with acetic acid (VIA), and HPV DNA testing, have limitations, requiring confirmation through colposcopy. This study introduces CerviCARE AI, an artificial intelligence (AI) analysis software, to address colposcopy challenges. It automatically analyzes Tele-cervicography images, distinguishing between low-grade and high-grade lesions. In a multicenter retrospective study, CerviCARE AI achieved a remarkable sensitivity of 98% for high-risk groups (P2, P3, HSIL or higher, CIN2 or higher) and a specificity of 95.5%. These findings underscore CerviCARE AI's potential as a valuable diagnostic tool for highly accurate identification of cervical precancerous lesions. While further prospective research is needed to validate its clinical utility, this AI system holds promise for improving cervical cancer screening and lessening the burden of this deadly disease.

Artificial intelligence for classification and detection of oral mucosa lesions on photographs: a systematic review and meta-analysis

OBJECTIVE

This study aimed to review and synthesize studies using artificial intelligence (AI) for classifying, detecting, or segmenting oral mucosal lesions on photographs.

MATERIALS AND METHOD

Inclusion criteria were (1) studies employing AI to (2) classify, detect, or segment oral mucosa lesions, (3) on oral photographs of human subjects. Included studies were assessed for risk of bias using Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2). A PubMed, Scopus, Embase, Web of Science, IEEE, arXiv, medRxiv, and grey literature (Google Scholar) search was conducted until June 2023, without language limitation.

RESULTS

After initial searching, 36 eligible studies (from 8734 identified records) were included. Based on QUADAS-2, only 7% of studies were at low risk of bias for all domains. Studies employed different AI models and reported a wide range of outcomes and metrics. The accuracy of AI for detecting oral mucosal lesions ranged from 74 to 100%, while that for clinicians un-aided by AI ranged from 61 to 98%. Pooled diagnostic odds ratio for studies which evaluated AI for diagnosing or discriminating potentially malignant lesions was 155 (95% confidence interval 23-1019), while that for cancerous lesions was 114 (59-221).

CONCLUSIONS

AI may assist in oral mucosa lesion screening while the expected accuracy gains or further health benefits remain unclear so far.

CLINICAL RELEVANCE

Artificial intelligence assists oral mucosa lesion screening and may foster more targeted testing and referral in the hands of non-specialist providers, for example. So far, it remains unclear if accuracy gains compared with specialized can be realized.

Artificial intelligence-based image analysis in clinical testing: lessons from cervical cancer screening

Novel screening and diagnostic tests based on artificial intelligence (AI) image recognition algorithms are proliferating. Some initial reports claim outstanding accuracy followed by disappointing lack of confirmation, including our own early work on cervical screening. This is a presentation of lessons learned, organized as a conceptual step-by-step approach to bridge the gap between the creation of an AI algorithm and clinical efficacy. The first fundamental principle is specifying rigorously what the algorithm is designed to identify and what the test is intended to measure (eg, screening, diagnostic, or prognostic). Second, designing the AI algorithm to minimize the most clinically important errors. For example, many equivocal cervical images cannot yet be labeled because the borderline between cases and controls is blurred. To avoid a misclassified case-control dichotomy, we have isolated the equivocal cases and formally included an intermediate, indeterminate class (severity order of classes: case>indeterminate>control). The third principle is evaluating AI algorithms like any other test, using clinical epidemiologic criteria. Repeatability of the algorithm at the borderline, for indeterminate images, has proven extremely informative. Distinguishing between internal and external validation is also essential. Linking the AI algorithm results to clinical risk estimation is the fourth principle. Absolute risk (not relative) is the critical metric for translating a test result into clinical use. Finally, generating risk-based guidelines for clinical use that match local resources and priorities is the last principle in our approach. We are particularly interested in applications to lower-resource settings to address health disparities. We note that similar principles apply to other domains of AI-based image analysis for medical diagnostic testing.

An automated cervical cancer diagnosis using genetic algorithm and CANFIS approaches

BACKGROUND

Cervical malignancy is considered among the most perilous cancers affecting women in numerous East African and South Asian nations, both in terms of its prevalence and fatality rates.

OBJECTIVE

This research aims to propose an efficient automated system for the segmentation of cancerous regions in cervical images.

METHODS

The proposed techniques encompass preprocessing, feature extraction with an optimized feature set, classification, and segmentation. The original cervical image undergoes smoothing using the Gaussian Filter technique, followed by the extraction of Local Binary Pattern (LBP) and Grey Level Co-occurrence Matrix (GLCM) features from the enhanced cervical images. LBP features capture pixel relationships within a mask window, while GLCM features quantify energy metrics across all pixels in the images. These features serve to distinguish normal cervical images from abnormal ones. The extracted features are optimized using Genetic Algorithm (GA) as an optimization method, and the optimized sets of features are classified using the Co-Active Adaptive Neuro-Fuzzy Inference System (CANFIS) classification method. Subsequently, a morphological segmentation technique is employed to categorize irregular cervical images, identifying and segmenting malignant regions within them.

RESULTS

The proposed approach achieved a sensitivity of 99.09%, specificity of 99.39%, and accuracy of 99.36%.

CONCLUSION

The proposed approach demonstrated superior performance compared to state-of-the-art techniques, and the results have been validated by expert radiologists.

Tritesting in Battor, Ghana: an integrated cervical precancer screening strategy to mitigate the challenges of multiple screening visits and loss to follow-up

Background

Human papillomavirus (HPV) DNA testing is more sensitive than cytology for detecting cervical precancer; however, increasing reports of high-risk HPV (hr-HPV)-negative cases of cervical intraepithelial neoplasia (CIN) and even malignancy motivate the use of combined testing. We present our experience with 'tritesting', defined as the performance of HPV DNA testing, cytology and visual inspection in a single session at the Cervical Cancer Prevention and Training Centre, Ghana. We further determined the prevalence rates of hr-HPV infection, abnormal cytology and cervical lesions among women screened using tritesting.

Methods

This descriptive retrospective cross-sectional study assessed all women screened via tritesting between April 2019 to April 2023. HPV DNA testing was performed using the Sansure MA-6000, GeneXpert or AmpFire platforms. Visual inspection was performed using enhanced visual assessment mobile colposcopy or visual inspection with acetic acid. Liquid-based cytology was performed using cervical samples taken with a Cervex-Brush® and fixed in PreservCyt, while samples for conventional cytology were taken using an Ayre spatula and cytobrush.

Results

Among 236 women screened (mean age, 39.1 years (standard deviation, 10.9)), the overall prevalence rates of hr-HPV infection and cervical lesions were 17.8% (95% confidence interval (CI), 13.1-23.3) and 11.9% (95% CI, 8.0-16.7), respectively. Cytology yielded findings of atypical squamous cells of undetermined significance or worse in 2.5% (95% CI, 0.9-5.5) of women. Histopathology following loop electrosurgical excision procedure revealed CIN I (tritest positive) and CIN III (hr-HPV-positive, visual inspection 'positive', cytology-negative) in one woman each. Factors independently associated with hr-HPV infection among 'tritested' women were age ≥ 39 years, tertiary level of education and current contraceptive use. Twenty-seven out of 39 hr-HPV-positive women (69.2%; 95% CI, 52.4-83.0) showed a type 3 transformation zone and would have needed to be recalled for a cytologic sample to be taken in a 'see and triage' approach with HPV DNA testing and a visual inspection method.

Conclusion

This study brings tritesting into the spotlight, as an alternative to other methods, particularly for women who prefer this due to the advantage of a single visit to a health facility and being more cost-effective, if they have to travel long distances to access cervical screening services.

Reproducible and clinically translatable deep neural networks for cervical screening

Cervical cancer is a leading cause of cancer mortality, with approximately 90% of the 250,000 deaths per year occurring in low- and middle-income countries (LMIC). Secondary prevention with cervical screening involves detecting and treating precursor lesions; however, scaling screening efforts in LMIC has been hampered by infrastructure and cost constraints. Recent work has supported the development of an artificial intelligence (AI) pipeline on digital images of the cervix to achieve an accurate and reliable diagnosis of treatable precancerous lesions. In particular, WHO guidelines emphasize visual triage of women testing positive for human papillomavirus (HPV) as the primary screen, and AI could assist in this triage task. In this work, we implemented a comprehensive deep-learning model selection and optimization study on a large, collated, multi-geography, multi-institution, and multi-device dataset of 9462 women (17,013 images). We evaluated relative portability, repeatability, and classification performance. The top performing model, when combined with HPV type, achieved an area under the Receiver Operating Characteristics (ROC) curve (AUC) of 0.89 within our study population of interest, and a limited total extreme misclassification rate of 3.4%, on held-aside test sets. Our model also produced reliable and consistent predictions, achieving a strong quadratic weighted kappa (QWK) of 0.86 and a minimal %2-class disagreement (% 2-Cl. D.) of 0.69%, between image pairs across women. Our work is among the first efforts at designing a robust, repeatable, accurate and clinically translatable deep-learning model for cervical screening.

A Survey on Artificial Intelligence in Pulmonary Imaging

Over the last decade, deep learning (DL) has contributed a paradigm shift in computer vision and image recognition creating widespread opportunities of using artificial intelligence in research as well as industrial applications. DL has been extensively studied in medical imaging applications, including those related to pulmonary diseases. Chronic obstructive pulmonary disease, asthma, lung cancer, pneumonia, and, more recently, COVID-19 are common lung diseases affecting nearly 7.4% of world population. Pulmonary imaging has been widely investigated toward improving our understanding of disease etiologies and early diagnosis and assessment of disease progression and clinical outcomes. DL has been broadly applied to solve various pulmonary image processing challenges including classification, recognition, registration, and segmentation. This paper presents a survey of pulmonary diseases, roles of imaging in translational and clinical pulmonary research, and applications of different DL architectures and methods in pulmonary imaging with emphasis on DL-based segmentation of major pulmonary anatomies such as lung volumes, lung lobes, pulmonary vessels, and airways as well as thoracic musculoskeletal anatomies related to pulmonary diseases.

Artificial intelligence-based risk stratification, accurate diagnosis and treatment prediction in gynecologic oncology

As data-driven science, artificial intelligence (AI) has paved a promising path toward an evolving health system teeming with thrilling opportunities for precision oncology. Notwithstanding the tremendous success of oncological AI in such fields as lung carcinoma, breast tumor and brain malignancy, less attention has been devoted to investigating the influence of AI on gynecologic oncology. Hereby, this review sheds light on the ever-increasing contribution of state-of-the-art AI techniques to the refined risk stratification and whole-course management of patients with gynecologic tumors, in particular, cervical, ovarian and endometrial cancer, centering on information and features extracted from clinical data (electronic health records), cancer imaging including radiological imaging, colposcopic images, cytological and histopathological digital images, and molecular profiling (genomics, transcriptomics, metabolomics and so forth). However, there are still noteworthy challenges beyond performance validation. Thus, this work further describes the limitations and challenges faced in the real-word implementation of AI models, as well as potential solutions to address these issues.

Artificial intelligence in breast cancer: application and future perspectives

Breast cancer is one of the most common cancers and is one of the leading causes of cancer-related deaths in women worldwide. Early diagnosis and treatment are the key for a favorable prognosis. The application of artificial intelligence technology in the medical field is increasingly extensive, including image analysis, automated diagnosis, intelligent pharmaceutical system, personalized treatment and so on. AI-based breast cancer imaging, pathology and adjuvant therapy technology cannot only reduce the workload of clinicians, but also continuously improve the accuracy and sensitivity of breast cancer diagnosis and treatment. This paper reviews the application of AI in breast cancer, as well as looks ahead and poses challenges to the future development of AI for breast cancer detection and therapeutic, so as to provide ideas for future research.

Artificial intelligence and visual inspection in cervical cancer screening

INTRODUCTION

Visual inspection with acetic acid is limited by subjectivity and a lack of skilled human resource. A decision support system based on artificial intelligence could address these limitations. We conducted a diagnostic study to assess the diagnostic performance using visual inspection with acetic acid under magnification of healthcare workers, experts, and an artificial intelligence algorithm.

METHODS

A total of 22 healthcare workers, 9 gynecologists/experts in visual inspection with acetic acid, and the algorithm assessed a set of 83 images from existing datasets with expert consensus as the reference. Their diagnostic performance was determined by analyzing sensitivity, specificity, and area under the curve, and intra- and inter-observer agreement was measured using Fleiss kappa values.

RESULTS

Sensitivity, specificity, and area under the curve were, respectively, 80.4%, 80.5%, and 0.80 (95% CI 0.70 to 0.90) for the healthcare workers, 81.6%, 93.5%, and 0.93 (95% CI 0.87 to 1.00) for the experts, and 80.0%, 83.3%, and 0.84 (95% CI 0.75 to 0.93) for the algorithm. Kappa values for the healthcare workers, experts, and algorithm were 0.45, 0.68, and 0.63, respectively.

CONCLUSION

This study enabled simultaneous assessment and demonstrated that expert consensus can be an alternative to histopathology to establish a reference standard for further training of healthcare workers and the artificial intelligence algorithm to improve diagnostic accuracy.

Implementing Smartphone-Based Telemedicine for Cervical Cancer Screening in Uganda: Qualitative Study of Stakeholders' Perceptions

BACKGROUND

In Uganda, cervical cancer (CaCx) is the commonest cancer, accounting for 35.7% of all cancer cases in women. The rates of human papillomavirus vaccination and CaCx screening remain low. Digital health tools and interventions have the potential to improve different aspects of CaCx screening and control in Uganda.

OBJECTIVE

This study aimed to describe stakeholders' perceptions of the telemedicine system we developed to improve CaCx screening in Uganda.

METHODS

We developed and implemented a smartphone-based telemedicine system for capturing and sharing cervical images and other clinical data, as well as an artificial intelligence model for automatic analysis of images. We conducted focus group discussions with health workers at the screening clinics (n=27) and women undergoing screening (n=15) to explore their perceptions of the system. The focus group discussions were supplemented with field observations and an evaluation survey of the health workers on system usability and the overall project.

RESULTS

In general, both patients and health workers had positive opinions about the system. Highlighted benefits included better cervical visualization, the ability to obtain a second opinion, improved communication between nurses and patients (to explain screening findings), improved clinical data management, performance monitoring and feedback, and modernization of screening service. However, there were also some negative perceptions. For example, some health workers felt the system is time-consuming, especially when it had just been introduced, while some patients were apprehensive about cervical image capture and sharing. Finally, commonplace challenges in digital health (eg, lack of interoperability and problems with sustainability) and challenges in cancer screening in general (eg, arduous referrals, inadequate monitoring and quality control) also resurfaced.

CONCLUSIONS

This study demonstrates the feasibility and value of digital health tools in CaCx screening in Uganda, particularly with regard to improving patient experience and the quality of screening services. It also provides examples of potential limitations that must be addressed for successful implementation.

Artificial intelligence in cancer diagnosis and therapy: Current status and future perspective

Artificial intelligence (AI) in healthcare plays a pivotal role in combating many fatal diseases, such as skin, breast, and lung cancer. AI is an advanced form of technology that uses mathematical-based algorithmic principles similar to those of the human mind for cognizing complex challenges of the healthcare unit. Cancer is a lethal disease with many etiologies, including numerous genetic and epigenetic mutations. Cancer being a multifactorial disease is difficult to be diagnosed at an early stage. Therefore, genetic variations and other leading factors could be identified in due time through AI and machine learning (ML). AI is the synergetic approach for mining the drug targets, their mechanism of action, and drug-organism interaction from massive raw data. This synergetic approach is also facing several challenges in data mining but computational algorithms from different scientific communities for multi-target drug discovery are highly helpful to overcome the bottlenecks in AI for drug-target discovery. AI and ML could be the epicenter in the medical world for the diagnosis, treatment, and evaluation of almost any disease in the near future. In this comprehensive review, we explore the immense potential of AI and ML when integrated with the biological sciences, specifically in the context of cancer research. Our goal is to illuminate the many ways in which AI and ML are being applied to the study of cancer, from diagnosis to individualized treatment. We highlight the prospective role of AI in supporting oncologists and other medical professionals in making informed decisions and improving patient outcomes by examining the intersection of AI and cancer control. Although AI-based medical therapies show great potential, many challenges must be overcome before they can be implemented in clinical practice. We critically assess the current hurdles and provide insights into the future directions of AI-driven approaches, aiming to pave the way for enhanced cancer interventions and improved patient care.

Automated Assessment of Digital Images of Uterine Cervix Captured Using Transvaginal Device-A Pilot Study

In low-resource settings, a point-of-care test for cervical cancer screening that can give an immediate result to guide management is urgently needed. A transvaginal digital device, "Smart Scope®" (SS), with an artificial intelligence-enabled auto-image-assessment (SS-AI) feature, was developed. In a single-arm observational study, eligible consenting women underwent a Smart Scope®-aided VIA-VILI test. Images of the cervix were captured using SS and categorized by SS-AI in four groups (green, amber, high-risk amber (HRA), red) based on risk assessment. Green and amber were classified as SS-AI negative while HRA and red were classified as SS-AI positive. The SS-AI-positive women were advised colposcopy and guided biopsy. The cervix images of SS-AI-negative cases were evaluated by an expert colposcopist (SS-M); those suspected of being positive were also recommended colposcopy and guided biopsy. Histopathology was considered a gold standard. Data on 877 SS-AI, 485 colposcopy, and 213 histopathology were available for analysis. The SS-AI showed high sensitivity (90.3%), specificity (75.3%), accuracy (84.04%), and correlation coefficient (0.670, p = 0.0) in comparison with histology at the CINI+ cutoff. In conclusion, the AI-enabled Smart Scope® test is a good alternative to the existing screening tests as it gives a real-time accurate assessment of cervical health and an opportunity for immediate triaging with visual evidence.

KD_ConvNeXt: knowledge distillation-based image classification of lung tumor surgical specimen sections

Introduction: Lung cancer is currently among the most prevalent and lethal cancers in the world in terms of incidence and fatality rates. In clinical practice, identifying the specific subtypes of lung cancer is essential in diagnosing and treating lung lesions. Methods: This paper aims to collect histopathological section images of lung tumor surgical specimens to construct a clinical dataset for researching and addressing the classification problem of specific subtypes of lung tumors. Our method proposes a teacher-student network architecture based on a knowledge distillation mechanism for the specific subtype classification of lung tumor histopathological section images to assist clinical applications, namely KD_ConvNeXt. The proposed approach enables the student network (ConvNeXt) to extract knowledge from the intermediate feature layers of the teacher network (Swin Transformer), improving the feature extraction and fitting capabilities of ConvNeXt. Meanwhile, Swin Transformer provides soft labels containing information about the distribution of images in various categories, making the model focused more on the information carried by types with smaller sample sizes while training. Results: This work has designed many experiments on a clinical lung tumor image dataset, and the KD_ConvNeXt achieved a superior classification accuracy of 85.64% and an F1-score of 0.7717 compared with other advanced image classification methods.

Enhancing Cervical Pre-Cancerous Classification Using Advanced Vision Transformer

One of the most common types of cancer among in women is cervical cancer. Incidence and fatality rates are steadily rising, particularly in developing nations, due to a lack of screening facilities, experienced specialists, and public awareness. Visual inspection is used to screen for cervical cancer after the application of acetic acid (VIA), histopathology test, Papanicolaou (Pap) test, and human papillomavirus (HPV) test. The goal of this research is to employ a vision transformer (ViT) enhanced with shifted patch tokenization (SPT) techniques to create an integrated and robust system for automatic cervix-type identification. A vision transformer enhanced with shifted patch tokenization is used in this work to learn the distinct features between the three different cervical pre-cancerous types. The model was trained and tested on 8215 colposcopy images of the three types, obtained from the publicly available mobile-ODT dataset. The model was tested on 30% of the whole dataset and it showed a good generalization capability of 91% accuracy. The state-of-the art comparison indicated the outperformance of our model. The experimental results show that the suggested system can be employed as a decision support tool in the detection of the cervical pre-cancer transformation zone, particularly in low-resource settings with limited experience and resources.

Pivotal Clinical Study to Evaluate the Efficacy and Safety of Assistive Artificial Intelligence-Based Software for Cervical Cancer Diagnosis

Colposcopy is the gold standard diagnostic tool for identifying cervical lesions. However, the accuracy of colposcopies depends on the proficiency of the colposcopist. Machine learning algorithms using an artificial intelligence (AI) system can quickly process large amounts of data and have been successfully applied in several clinical situations. This study evaluated the feasibility of an AI system as an assistive tool for diagnosing high-grade cervical intraepithelial neoplasia lesions compared to the human interpretation of cervical images. This two-centered, crossover, double-blind, randomized controlled trial included 886 randomly selected images. Four colposcopists (two proficient and two inexperienced) independently evaluated cervical images, once with and the other time without the aid of the Cerviray AI^® system (AIDOT, Seoul, Republic of Korea). The AI aid demonstrated improved areas under the curve on the localization receiver-operating characteristic curve compared with the colposcopy impressions of colposcopists (difference 0.12, 95% confidence interval, 0.10-0.14, p < 0.001). Sensitivity and specificity also improved when using the AI system (89.18% vs. 71.33%; p < 0.001, 96.68% vs. 92.16%; p < 0.001, respectively). Additionally, the classification accuracy rate improved with the aid of AI (86.40% vs. 75.45%; p < 0.001). Overall, the AI system could be used as an assistive diagnostic tool for both proficient and inexperienced colposcopists in cervical cancer screenings to estimate the impression and location of pathologic lesions. Further utilization of this system could help inexperienced colposcopists confirm where to perform a biopsy to diagnose high-grade lesions.

Strategies to eliminate cervical cancer in China

Cervical cancer is a widely distributed disease that is preventable and controllable through early intervention. The World Health Organization has identified three key measures, coverage populations and coverage targets to eliminate cervical cancer. The WHO and several countries have conducted model predictions to determine the optimal strategy and timing of cervical cancer elimination. However, specific implementation strategies need to be developed in the context of local conditions. China has a relatively high disease burden of cervical cancer but a low human papillomavirus vaccination rate and cervical cancer screening population coverage. The purpose of this paper is to review interventions and prediction studies for the elimination of cervical cancer and to analyze the problems, challenges and strategies for the elimination of cervical cancer in China.

Diagnosis Assistance in Colposcopy by Segmenting Acetowhite Epithelium Using U-Net with Images before and after Acetic Acid Solution Application

Colposcopy is an essential examination tool to identify cervical intraepithelial neoplasia (CIN), a precancerous lesion of the uterine cervix, and to sample its tissues for histological examination. In colposcopy, gynecologists visually identify the lesion highlighted by applying an acetic acid solution to the cervix using a magnifying glass. This paper proposes a deep learning method to aid the colposcopic diagnosis of CIN by segmenting lesions. In this method, to segment the lesion effectively, the colposcopic images taken before acetic acid solution application were input to the deep learning network, U-Net, for lesion segmentation with the images taken following acetic acid solution application. We conducted experiments using 30 actual colposcopic images of acetowhite epithelium, one of the representative types of CIN. As a result, it was confirmed that accuracy, precision, and F1 scores, which were 0.894, 0.837, and 0.834, respectively, were significantly better when images taken before and after acetic acid solution application were used than when only images taken after acetic acid solution application were used (0.882, 0.823, and 0.823, respectively). This result indicates that the image taken before acetic acid solution application is helpful for accurately segmenting the CIN in deep learning.

Use of risk-based cervical screening programs in resource-limited settings

Cervical cancer screening and management in the U.S. has adopted a risk-based approach. However, the majority of cervical cancer cases and deaths occur in resource-limited settings, where screening and management are not widely available. We describe a conceptual model that optimizes cervical cancer screening and management in resource-limited settings by utilizing a risk-based approach. The principles of risk-based screening and management in resource limited settings include (1) ensure that the screening method effectively separates low-risk from high-risk patients; (2) directing resources to populations at the highest cancer risk; (3) screen using HPV testing via self-sampling; (4) utilize HPV genotyping to improve risk stratification and better determine who will benefit from treatment, and (5) automated visual evaluation with artificial intelligence may further improve risk stratification. Risk-based screening and management in resource limited settings can optimize prevention by focusing triage and treatment resources on the highest risk patients while minimizing interventions in lower risk patients.

Deep Cervix Model Development from Heterogeneous and Partially Labeled Image Datasets

Cervical cancer is a significant disease affecting women worldwide. Regular cervical examination with gynecologists is important for early detection and treatment planning for women with precancers. Precancer is the direct precursor to cervical cancer. However, there is a scarcity of experts and the experts' assessments are subject to variations in interpretation. In this scenario, the development of a robust automated cervical image classification system is important to augment the experts' limitations. Ideally, for such a system the class label prediction will vary according to the cervical inspection objectives. Hence, the labeling criteria may not be the same in the cervical image datasets. Moreover, due to the lack of confirmatory test results and inter-rater labeling variation, many images are left unlabeled. Motivated by these challenges, we propose to develop a pretrained cervix model from heterogeneous and partially labeled cervical image datasets. Self-supervised Learning (SSL) is employed to build the cervical model. Further, considering data-sharing restrictions, we show how federated self-supervised learning (FSSL) can be employed to develop a cervix model without sharing the cervical images. The task-specific classification models are developed by fine-tuning the cervix model. Two partially labeled cervical image datasets labeled with different classification criteria are used in this study. According to our experimental study, the cervix model prepared with dataset-specific SSL boosts classification accuracy by 2.5%↑ than ImageNet pretrained model. The classification accuracy is further boosted by 1.5%↑ when images from both datasets are combined for SSL. We see that in comparison with the dataset-specific cervix model developed with SSL, the FSSL is performing better.

Machine Learning and Explainable Artificial Intelligence Using Counterfactual Explanations for Evaluating Posture Parameters

Postural deficits such as hyperlordosis (hollow back) or hyperkyphosis (hunchback) are relevant health issues. Diagnoses depend on the experience of the examiner and are, therefore, often subjective and prone to errors. Machine learning (ML) methods in combination with explainable artificial intelligence (XAI) tools have proven useful for providing an objective, data-based orientation. However, only a few works have considered posture parameters, leaving the potential for more human-friendly XAI interpretations still untouched. Therefore, the present work proposes an objective, data-driven ML system for medical decision support that enables especially human-friendly interpretations using counterfactual explanations (CFs). The posture data for 1151 subjects were recorded by means of stereophotogrammetry. An expert-based classification of the subjects regarding the presence of hyperlordosis or hyperkyphosis was initially performed. Using a Gaussian progress classifier, the models were trained and interpreted using CFs. The label errors were flagged and re-evaluated using confident learning. Very good classification performances for both hyperlordosis and hyperkyphosis were found, whereby the re-evaluation and correction of the test labels led to a significant improvement (M_PRAUC = 0.97). A statistical evaluation showed that the CFs seemed to be plausible, in general. In the context of personalized medicine, the present study's approach could be of importance for reducing diagnostic errors and thereby improving the individual adaptation of therapeutic measures. Likewise, it could be a basis for the development of apps for preventive posture assessment.

Performance of visual inspection of the cervix with acetic acid (VIA) for triage of HPV screen-positive women: results from the ESTAMPA study

VIA is recommended for triage of HPV-positive women attending cervical screening. In the multicentric ESTAMPA study, VIA performance for detection of cervical intraepithelial neoplasia grade 3 or worse (CIN3+) among HPV-positive women was evaluated. Women aged 30-64 years were screened with HPV testing and cytology and referred to colposcopy if either test was positive. At colposcopy visit, study-trained midwives/nurses/GPs performed VIA ahead of colposcopy. VIA was considered positive if acetowhite lesions were observed in or close to the transformation zone. Ablative treatment eligibility was assessed for VIA positives. Performance indicators were estimated. Three thousand one hundred and forty-two HPV-positive women were included. Sensitivity for CIN3+ was 85.9% (95% CI 81.2-89.5) among women <50 years and, although not significant, slightly lower in women 50+ (78.0%, 95% CI 65.9-86.6). Overall specificity was 58.6% (95% CI 56.7-60.5) and was significantly higher among women 50+ (70.3%, 95% CI 66.8-73.5) compared to women <50 (54.3%, 95% CI 52.1-56.5). VIA positivity was lower among women 50+ (35.2%, 95% CI 31.9-38.6) compared to women <50 (53.2, 95% CI 51.1-55.2). Overall eligibility for ablative treatment was 74.5% and did not differ by age. VIA sensitivity, specificity, and positivity, and ablative treatment eligibility varied highly by provider (ranges: 25%-95.4%, 44.9%-94.4%, 8.2%-65.3%, 0%-98.7%, respectively). VIA sensitivity for cervical precancer detection among HPV-positive women performed by trained providers was high with an important reduction in referral rates. However, scaling-up HPV screening triaged by VIA will be challenging due to the high variability of VIA performance and providers' need for training and supervision.

Anatomically constrained and attention-guided deep feature fusion for joint segmentation and deformable medical image registration

The main objective of anatomically plausible results for deformable image registration is to improve model's registration accuracy by minimizing the difference between a pair of fixed and moving images. Since many anatomical features are closely related to each other, leveraging supervision from auxiliary tasks (such as supervised anatomical segmentation) has the potential to enhance the realism of the warped images after registration. In this work, we employ a Multi-Task Learning framework to formulate registration and segmentation as a joint issue, in which we utilize anatomical constraint from auxiliary supervised segmentation to enhance the realism of the predicted images. First, we propose a Cross-Task Attention Block to fuse the high-level feature from both the registration and segmentation network. With the help of initial anatomical segmentation, the registration network can benefit from learning the task-shared feature correlation and rapidly focusing on the parts that need deformation. On the other hand, the anatomical segmentation discrepancy from ground-truth fixed annotations and predicted segmentation maps of initial warped images are integrated into the loss function to guide the convergence of the registration network. Ideally, a good deformation field should be able to minimize the loss function of registration and segmentation. The voxel-wise anatomical constraint inferred from segmentation helps the registration network to reach a global optimum for both deformable and segmentation learning. Both networks can be employed independently during the testing phase, enabling only the registration output to be predicted when the segmentation labels are unavailable. Qualitative and quantitative results indicate that our proposed methodology significantly outperforms the previous state-of-the-art approaches on inter-patient brain MRI registration and pre- and intra-operative uterus MRI registration tasks within our specific experimental setup, which leads to state-of-the-art registration quality scores of 0.755 and 0.731 (i.e., by 0.8% and 0.5% increases) DSC for both tasks, respectively.

Artificial intelligence in colposcopic examination: A promising tool to assist junior colposcopists

Introduction

Well-trained colposcopists are in huge shortage worldwide, especially in low-resource areas. Here, we aimed to evaluate the Colposcopic Artificial Intelligence Auxiliary Diagnostic System (CAIADS) to detect abnormalities based on digital colposcopy images, especially focusing on its role in assisting junior colposcopist to correctly identify the lesion areas where biopsy should be performed.

Materials and methods

This is a hospital-based retrospective study, which recruited the women who visited colposcopy clinics between September 2021 to January 2022. A total of 366 of 1,146 women with complete medical information recorded by a senior colposcopist and valid histology results were included. Anonymized colposcopy images were reviewed by CAIADS and a junior colposcopist separately, and the junior colposcopist reviewed the colposcopy images with CAIADS results (named CAIADS-Junior). The diagnostic accuracy and biopsy efficiency of CAIADS and CAIADS-Junior were assessed in detecting cervical intraepithelial neoplasia grade 2 or worse (CIN2+), CIN3+, and cancer in comparison with the senior and junior colposcipists. The factors influencing the accuracy of CAIADS were explored.

Results

For CIN2 + and CIN3 + detection, CAIADS showed a sensitivity at ~80%, which was not significantly lower than the sensitivity achieved by the senior colposcopist (for CIN2 +: 80.6 vs. 91.3%, p = 0.061 and for CIN3 +: 80.0 vs. 90.0%, p = 0.189). The sensitivity of the junior colposcopist was increased significantly with the assistance of CAIADS (for CIN2 +: 95.1 vs. 79.6%, p = 0.002 and for CIN3 +: 97.1 vs. 85.7%, p = 0.039) and was comparable to those of the senior colposcopists (for CIN2 +: 95.1 vs. 91.3%, p = 0.388 and for CIN3 +: 97.1 vs. 90.0%, p = 0.125). In detecting cervical cancer, CAIADS achieved the highest sensitivity at 100%. For all endpoints, CAIADS showed the highest specificity (55–64%) and positive predictive values compared to both senior and junior colposcopists. When CIN grades became higher, the average biopsy numbers decreased for the subspecialists and CAIADS required a minimum number of biopsies to detect per case (2.2–2.6 cut-points). Meanwhile, the biopsy sensitivity of the junior colposcopist was the lowest, but the CAIADS-assisted junior colposcopist achieved a higher biopsy sensitivity.

Conclusion

Colposcopic Artificial Intelligence Auxiliary Diagnostic System could assist junior colposcopists to improve diagnostic accuracy and biopsy efficiency, which might be a promising solution to improve the quality of cervical cancer screening in low-resource settings.

ColpoClassifier: A Hybrid Framework for Classification of the Cervigrams

Colposcopy plays a vital role in detecting cervical cancer. Artificial intelligence-based methods have been implemented in the literature for the classification of colposcopy images. However, there is a need for a more effective method that can accurately classify cervigrams. In this paper, ColpoClassifier, a hybrid framework for the classification of cervigrams, is proposed, which consists of feature extraction followed by classification. This paper uses a Gray-level co-occurrence matrix (GLCM), a Gray-level run length matrix (GLRLM), and a histogram of gradients (HOG) for feature extraction. These features are combined to form a feature fusion vector of the form GLCM + GLRLM + HOG. The different machine learning classifiers are used for classification by using individual feature vectors as well as feature fusion vectors. The dataset used in this paper is compiled by downloading images from the WHO website. Two variants of this dataset are created, Dataset-I contains images of the aceto-whitening effect, green filter, iodine application, and raw cervigram while Dataset-II only contains images of the aceto-whitening effect. This paper presents the classification performance on all kinds of images with the individual as well as hybrid feature fusion vector and concludes that hybrid feature fusion vectors on aceto-whitening images have given the best results.

REPRODUCIBLE AND CLINICALLY TRANSLATABLE DEEP NEURAL NETWORKS FOR CANCER SCREENING

Cervical cancer is a leading cause of cancer mortality, with approximately 90% of the 250,000 deaths per year occurring in low- and middle-income countries (LMIC). Secondary prevention with cervical screening involves detecting and treating precursor lesions; however, scaling screening efforts in LMIC has been hampered by infrastructure and cost constraints. Recent work has supported the development of an artificial intelligence (AI) pipeline on digital images of the cervix to achieve an accurate and reliable diagnosis of treatable precancerous lesions. In particular, WHO guidelines emphasize visual triage of women testing positive for human papillomavirus (HPV) as the primary screen, and AI could assist in this triage task. Published AI reports have exhibited overfitting, lack of portability, and unrealistic, near-perfect performance estimates. To surmount recognized issues, we implemented a comprehensive deep-learning model selection and optimization study on a large, collated, multi-institutional dataset of 9,462 women (17,013 images). We evaluated relative portability, repeatability, and classification performance. The top performing model, when combined with HPV type, achieved an area under the Receiver Operating Characteristics (ROC) curve (AUC) of 0.89 within our study population of interest, and a limited total extreme misclassification rate of 3.4%, on held-aside test sets. Our work is among the first efforts at designing a robust, repeatable, accurate and clinically translatable deep-learning model for cervical screening.

A novel patches-selection method for the classification of point-of-care biosensing lateral flow assays with cardiac biomarkers

Cardiovascular Disease (CVD) is amongst the leading cause of death globally, which calls for rapid detection and treatment. Biosensing devices are used for the diagnosis of cardiovascular disease at the point-of-care (POC), with lateral flow assays (LFAs) being particularly useful. However, due to their low sensitivity, most LFAs have been shown to have difficulties detecting low analytic concentrations. Breakthroughs in artificial intelligence (AI) and image processing reduced this detection constraint and improved disease diagnosis. This paper presents a novel patches-selection approach for generating LFA images from the test line and control line of LFA images, analyzing the image features, and utilizing them to reliably predict and classify LFA images by deploying classification algorithms, specifically Convolutional Neural Networks (CNNs). The generated images were supplied as input data to the CNN model, a strong model for extracting crucial information from images, to classify the target images and provide risk stratification levels to medical professionals. With this approach, the classification model produced about 98% accuracy, and as per the literature review, this approach has not been investigated previously. These promising results show the proposed method may be useful for identifying a wide variety of diseases and conditions, including cardiovascular problems.

Colposcopic accuracy in diagnosing squamous intraepithelial lesions: a systematic review and meta-analysis of the International Federation of Cervical Pathology and Colposcopy 2011 terminology

BACKGROUND

Colposcopy is an important tool in diagnosing cervical cancer, and the International Federation of Cervical Pathology and Colposcopy (IFCPC) issued the latest version of the guidelines in 2011. This study aims to systematically assess the accuracy of colposcopy in predicting low-grade squamous intraepithelial lesions or worse (LSIL+) / high-grade squamous intraepithelial lesions or worse (HSIL+) under the 2011 IFCPC terminology.

METHODS

We performed a systematic review and meta-analysis, following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. We searched for studies about the performance of colposcopy in diagnosing cervical intraepithelial neoplasia under the new IFCPC colposcopy terminology from PubMed, Embase, Web of Science and the Cochrane database. Data were independently extracted by two authors and an overall diagnostic performance index was calculated under two colposcopic thresholds.

RESULTS

Totally, fifteen articles with 22,764 participants in compliance with the criteria were included in meta-analysis. When colposcopy was used to detect LSIL+, the combined sensitivity and specificity were 0.92 (95% CI 0.88-0.95) and 0.51 (0.43-0.59), respectively. When colposcopy was used to detect HSIL+, the combined sensitivity and specificity were 0.68 (0.58-0.76) and 0.93 (0.88-0.96), respectively.

CONCLUSION

In accordance with the 2011 IFCPC terminology, the accuracy of colposcopy has improved in terms of both sensitivity and specificity. Colposcopy is now more sensitive with LSIL+ taken as the cut-off value and is more specific to HSIL+. These findings suggest we are avoiding under- or overdiagnosis both of which impact on patients' well-being.

Artificial Intelligence-Based Cervical Cancer Screening on Images Taken during Visual Inspection with Acetic Acid: A Systematic Review

Visual inspection with acetic acid (VIA) is one of the methods recommended by the World Health Organization for cervical cancer screening. VIA is simple and low-cost; it, however, presents high subjectivity. We conducted a systematic literature search in PubMed, Google Scholar and Scopus to identify automated algorithms for classifying images taken during VIA as negative (healthy/benign) or precancerous/cancerous. Of the 2608 studies identified, 11 met the inclusion criteria. The algorithm with the highest accuracy in each study was selected, and some of its key features were analyzed. Data analysis and comparison between the algorithms were conducted, in terms of sensitivity and specificity, ranging from 0.22 to 0.93 and 0.67 to 0.95, respectively. The quality and risk of each study were assessed following the QUADAS-2 guidelines. Artificial intelligence-based cervical cancer screening algorithms have the potential to become a key tool for supporting cervical cancer screening, especially in settings where there is a lack of healthcare infrastructure and trained personnel. The presented studies, however, assess their algorithms using small datasets of highly selected images, not reflecting whole screened populations. Large-scale testing in real conditions is required to assess the feasibility of integrating those algorithms in clinical settings.

Comparison of human papillomavirus-based cervical cancer screening strategies in Tanzania among women with and without HIV

Cervical cancer is the most common female cancer in Eastern Africa, and the World Health Organization (WHO) recommends human papillomavirus (HPV)-based screening as a key element to eliminate the disease. In this cross-sectional study from Tanzania, we compared nine HPV-based cervical cancer screening strategies, including HPV testing at standard cut-off; HPV testing at increased viral load cut-offs; HPV testing with partial/extended genotyping, and HPV testing with visual inspection with acetic acid (VIA). We pooled data collected during 2008 to 2009 and 2015 to 2017 from 6851 women aged 25 to 65. Cervical cytology samples were HPV tested with Hybrid Capture 2, and HPV positive samples were genotyped with INNO-LiPA Extra II. Human immunodeficiency virus (HIV) testing and VIA were done according to local standards. We calculated sensitivity, specificity, positive and negative predictive value of screening strategies, with high-grade cytological lesions as reference, separately for women with and without HIV. HPV testing at standard cut-off (1.0 relative light units [RLU]) had highest sensitivity (HIV+: 97.8%; HIV-: 91.5%), but moderate specificity (HIV+: 68.1%; HIV-: 85.7%). Increasing the cut-off for HPV positivity to higher viral loads (5.0/10.0 RLU) increased specificity (HIV+: 74.2%-76.5%; HIV-: 89.5%-91.2%), with modest sensitivity reductions (HIV+: 91.3%-95.7%; HIV-: 83.5%-87.8%). Limiting test positivity to HPV types 16/18/31/33/35/45/52/58 improved specificity while maintaining high sensitivity (HIV+: 90.2%; HIV-: 81.1%). Triage with VIA and/or partial genotyping for HPV16/18 or HPV16/18/45 had low sensitivities (≤65%). In conclusion, HPV testing alone, or HPV testing with extended genotyping or increased viral load cut-offs, may improve cervical cancer screening in Sub-Saharan Africa.

Prevention Strategies and Early Diagnosis of Cervical Cancer: Current State and Prospects

Cervical cancer ranks third among all new cancer cases and causes of cancer deaths in females. The paper provides an overview of cervical cancer prevention strategies employed in different regions, with incidence and mortality rates ranging from high to low. It assesses the effectiveness of approaches proposed by national healthcare systems by analysing data published in the National Library of Medicine (Pubmed) since 2018 featuring the following keywords: "cervical cancer prevention", "cervical cancer screening", "barriers to cervical cancer prevention", "premalignant cervical lesions" and "current strategies". WHO's 90-70-90 global strategy for cervical cancer prevention and early screening has proven effective in different countries in both mathematical models and clinical practice. The data analysis carried out within this study identified promising approaches to cervical cancer screening and prevention, which can further enhance the effectiveness of the existing WHO strategy and national healthcare systems. One such approach is the application of AI technologies for detecting precancerous cervical lesions and choosing treatment strategies. As such studies show, the use of AI can not only increase detection accuracy but also ease the burden on primary care.

Implementation of visual inspection with acetic acid and Lugol's iodine for cervical cancer screening in rural China

OBJECTIVE

To address the value of visual inspection where HPV-based screening is not yet available, we evaluated the real-world effectiveness of visual inspection with acetic acid (VIA) and with Lugol's iodine (VILI) as a primary screening method for cervical cancer in rural China.

METHODS

A total of 206 133 women aged 30-59 years received two rounds of VIA/VILI screening for cervical cancer in 2006-2010. Women with positive screening results underwent colposcopy and direct biopsy, and were treated if cervical intraepithelial neoplasia grade 2 or worse (CIN2+) was diagnosed. Clinical effectiveness of VIA/VILI was evaluated by process and outcome measures.

RESULTS

The VIA/VILI positivity rate, biopsy rate and detection rate of CIN2+ in the second round were significantly lower than in the first round. The 2-year cumulative detection rate of CIN2+ varied from 0.53% to 0.90% among the four cohorts initiated in 2006, 2007, 2008, and 2009. The first round of screening detected 60%-83% of CIN2, 70%-86% of CIN3, and 88%-100% of cervical cancer. Over 92% of CIN2+ were found at the early stage.

CONCLUSION

Multiple rounds of visual inspection with continuous training and quality assurance could act as a temporary substitutional screening method for cervical cancer in resource-restricted settings.

Automated In Vivo High-Resolution Imaging to Detect Human Papillomavirus-Associated Anal Precancer in Persons Living With HIV

INTRODUCTION

In the United States, the effectiveness of anal cancer screening programs has been limited by a lack of trained professionals proficient in high-resolution anoscopy (HRA) and a high patient lost-to-follow-up rate between diagnosis and treatment. Simplifying anal intraepithelial neoplasia grade 2 or more severe (AIN 2+) detection could radically improve the access and efficiency of anal cancer prevention. Novel optical imaging providing point-of-care diagnoses could substantially improve existing HRA and histology-based diagnosis. This work aims to demonstrate the potential of high-resolution microendoscopy (HRME) coupled with a novel machine learning algorithm for the automated, in vivo diagnosis of anal precancer.

METHODS

The HRME, a fiber-optic fluorescence microscope, was used to capture real-time images of anal squamous epithelial nuclei. Nuclear staining is achieved using 0.01% wt/vol proflavine, a topical contrast agent. HRME images were analyzed by a multitask deep learning network (MTN) that computed the probability of AIN 2+ for each HRME image.

RESULTS

The study accrued data from 77 people living with HIV. The MTN achieved an area under the receiver operating curve of 0.84 for detection of AIN 2+. At the AIN 2+ probability cutoff of 0.212, the MTN achieved comparable performance to expert HRA impression with a sensitivity of 0.92 ( P = 0.68) and specificity of 0.60 ( P = 0.48) when using histopathology as the gold standard.

DISCUSSION

When used in combination with HRA, this system could facilitate more selective biopsies and promote same-day AIN2+ treatment options by enabling real-time diagnosis.

Using deep learning to predict survival outcome in non-surgical cervical cancer patients based on pathological images

PURPOSE

We analyzed clinical features and the representative HE-stained pathologic images to predict 5-year overall survival via the deep-learning approach in cervical cancer patients in order to assist oncologists in designing the optimal treatment strategies.

METHODS

The research retrospectively collected 238 non-surgical cervical cancer patients treated with radiochemotherapy from 2014 to 2017. These patients were randomly divided into the training set (n = 165) and test set (n = 73). Then, we extract deep features after segmenting the HE-stained image into patches of size 224 × 224. A Lasso-Cox model was constructed with clinical data to predict 5-year OS. C-index evaluated this model performance with 95% CI, calibration curve, and ROC.

RESULTS

Based on multivariate analysis, 2 of 11 clinical characteristics (C-index 0.68) and 2 of 2048 pathomic features (C-index 0.74) and clinical-pathomic model (C-index 0.83) of nomograms predict 5-year survival in the training set, respectively. In test set, compared with the pathomic and clinical characteristics used alone, the clinical-pathomic model had an AUC of 0.750 (95% CI 0.540-0.959), the clinical predictor model had an AUC of 0.729 (95% CI 0.551-0.909), and the pathomic model AUC was 0.703 (95% CI 0.487-0.919). Based on appropriate nomogram scores, we divided patients into high-risk and low-risk groups, and Kaplan-Meier survival probability curves for both groups showed statistical differences.

CONCLUSION

We built a clinical-pathomic model to predict 5-year OS in non-surgical cervical cancer patients, which may be a promising method to improve the precision of personalized therapy.

Application of EfficientNet-B0 and GRU-based deep learning on classifying the colposcopy diagnosis of precancerous cervical lesions

BACKGROUND

Colposcopy is indispensable for the diagnosis of cervical lesions. However, its diagnosis accuracy for high-grade squamous intraepithelial lesion (HSIL) is at about 50%, and the accuracy is largely dependent on the skill and experience of colposcopists. The advancement in computational power made it possible for the application of artificial intelligence (AI) to clinical problems. Here, we explored the feasibility and accuracy of the application of AI on precancerous and cancerous cervical colposcopic image recognition and classification.

METHODS

The images were collected from 6002 colposcopy examinations of normal control, low-grade squamous intraepithelial lesion (LSIL), and HSIL. For each patient, the original, Schiller test, and acetic-acid images were all collected. We built a new neural network classification model based on the hybrid algorithm. EfficientNet-b0 was used as the backbone network for the image feature extraction, and GRU(Gate Recurrent Unit)was applied for feature fusion of the three modes examinations (original, acetic acid, and Schiller test).

RESULTS

The connected network classifier achieved an accuracy of 90.61% in distinguishing HSIL from normal and LSIL. Furthermore, the model was applied to "Trichotomy", which reached an accuracy of 91.18% in distinguishing the HSIL, LSIL and normal control at the same time.

CONCLUSION

Our results revealed that as shown by the high accuracy of AI in the classification of colposcopic images, AI exhibited great potential to be an effective tool for the accurate diagnosis of cervical disease and for early therapeutic intervention in cervical precancer.

Current developments in artificial intelligence from obstetrics and gynecology to urogynecology

In today's medical practice clinicians need to struggle with a huge amount of data to improve the outcomes of the patients. Sometimes one clinician needs to deal with thousands of ultrasound images or hundred papers of laboratory results. To overcome this shortage, computers get in help of human beings and they are educated under the term "artificial intelligence." We were using artificial intelligence in our daily lives (i.e., Google, Netflix, etc.), but applications in medicine are relatively new. In obstetrics and gynecology, artificial intelligence models mostly use ultrasound images for diagnostic purposes but nowadays researchers started to use other medical recordings like non-stress tests or urodynamics study results to develop artificial intelligence applications. Urogynecology is a developing subspecialty of obstetrics and gynecology, and articles about artificial intelligence in urogynecology are limited but in this review, we aimed to increase clinicians' knowledge about this new approach.

A deep learning-based method for cervical transformation zone classification in colposcopy images

BACKGROUND

Colposcopy is one of the common methods of cervical cancer screening. The type of cervical transformation zone is considered one of the important factors for grading colposcopic findings and choosing treatment.

OBJECTIVE

This study aims to develop a deep learning-based method for automatic classification of cervical transformation zone from colposcopy images.

METHODS

We proposed a multiscale feature fusion classification network to classify cervical transformation zone, which can extract features from images and fuse them at multiple scales. Cervical regions were first detected from original colposcopy images and then fed into our multiscale feature fusion classification network.

RESULTS

The results on the test dataset showed that, compared with the state-of-the-art image classification models, the proposed classification network had the highest classification accuracy, reaching 88.49%, and the sensitivity to type 1, type 2 and type 3 were 90.12%, 85.95% and 89.45%, respectively, higher than the comparison methods.

CONCLUSIONS

The proposed method can automatically classify cervical transformation zone in colposcopy images, and can be used as an auxiliary tool in cervical cancer screening.

Cervical pre-cancerous lesion detection: development of smartphone-based VIA application using artificial intelligence

OBJECTIVE

Visual inspection of cervix after acetic acid application (VIA) has been considered an alternative to Pap smear in resource-limited settings, like Indonesia. However, VIA results mainly depend on examiner's experience and with the lack of comprehensive training of healthcare workers, VIA accuracy keeps declining. We aimed to develop an artificial intelligence (AI)-based Android application that can automatically determine VIA results in real time and may be further developed as a health care support system in cervical cancer screening.

RESULT

A total of 199 women who underwent VIA test was studied. Images of cervix before and after VIA test were taken with smartphone, then evaluated and labelled by experienced oncologist as VIA positive or negative. Our AI model training pipeline consists of 3 steps: image pre-processing, feature extraction, and classifier development. Out of the 199 data, 134 were used as train-validation data and the remaining 65 data were used as test data. The trained AI model generated a sensitivity of 80%, specificity of 96.4%, accuracy of 93.8%, precision of 80%, and ROC/AUC of 0.85 (95% CI 0.66-1.0). The developed AI-based Android application may potentially aid cervical cancer screening, especially in low resource settings.

Deep learning-assisted sensitive detection of fentanyl using a bubbling-microchip

Deep learning-enabled smartphone-based image processing has significant advantages in the development of point-of-care diagnostics. Conventionally, most deep-learning applications require task specific large scale expertly annotated datasets. Therefore, these algorithms are oftentimes limited only to applications that have large retrospective datasets available for network development. Here, we report the possibility of utilizing adversarial neural networks to overcome this challenge by expanding the utility of non-specific data for the development of deep learning models. As a clinical model, we report the detection of fentanyl, a small molecular weight drug that is a type of opioid, at the point-of-care using a deep-learning empowered smartphone assay. We used the catalytic property of platinum nanoparticles (PtNPs) in a smartphone-enabled microchip bubbling assay to achieve high analytical sensitivity (detecting fentanyl at concentrations as low as 0.23 ng mL-1 in phosphate buffered saline (PBS), 0.43 ng mL-1 in human serum and 0.64 ng mL-1 in artificial human urine). Image-based inferences were made by our adversarial-based SPyDERMAN network that was developed using a limited dataset of 104 smartphone images of microchips with bubble signals from tests performed with known fentanyl concentrations and using our retrospective library of 17 573 non-specific bubbling-microchip images. The accuracy (± standard error of mean) of the developed system in determining the presence of fentanyl, when using a cutoff concentration of 1 ng mL-1, was 93 ± 0% in human serum (n = 100) and 95.3 ± 1.5% in artificial human urine (n = 100).

Improving the repeatability of deep learning models with Monte Carlo dropout

The integration of artificial intelligence into clinical workflows requires reliable and robust models. Repeatability is a key attribute of model robustness. Ideal repeatable models output predictions without variation during independent tests carried out under similar conditions. However, slight variations, though not ideal, may be unavoidable and acceptable in practice. During model development and evaluation, much attention is given to classification performance while model repeatability is rarely assessed, leading to the development of models that are unusable in clinical practice. In this work, we evaluate the repeatability of four model types (binary classification, multi-class classification, ordinal classification, and regression) on images that were acquired from the same patient during the same visit. We study the each model’s performance on four medical image classification tasks from public and private datasets: knee osteoarthritis, cervical cancer screening, breast density estimation, and retinopathy of prematurity. Repeatability is measured and compared on ResNet and DenseNet architectures. Moreover, we assess the impact of sampling Monte Carlo dropout predictions at test time on classification performance and repeatability. Leveraging Monte Carlo predictions significantly increases repeatability, in particular at the class boundaries, for all tasks on the binary, multi-class, and ordinal models leading to an average reduction of the 95% limits of agreement by 16% points and of the class disagreement rate by 7% points. The classification accuracy improves in most settings along with the repeatability. Our results suggest that beyond about 20 Monte Carlo iterations, there is no further gain in repeatability. In addition to the higher test-retest agreement, Monte Carlo predictions are better calibrated which leads to output probabilities reflecting more accurately the true likelihood of being correctly classified.

Diagnosis of Cervical Cancer and Pre-Cancerous Lesions by Artificial Intelligence: A Systematic Review

OBJECTIVE

The likelihood of timely treatment for cervical cancer increases with timely detection of abnormal cervical cells. Automated methods of detecting abnormal cervical cells were established because manual identification requires skilled pathologists and is time consuming and prone to error. The purpose of this systematic review is to evaluate the diagnostic performance of artificial intelligence (AI) technologies for the prediction, screening, and diagnosis of cervical cancer and pre-cancerous lesions.

MATERIALS AND METHODS

Comprehensive searches were performed on three databases: Medline, Web of Science Core Collection (Indexes = SCI-EXPANDED, SSCI, A & HCI Timespan) and Scopus to find papers published until July 2022. Articles that applied any AI technique for the prediction, screening, and diagnosis of cervical cancer were included in the review. No time restriction was applied. Articles were searched, screened, incorporated, and analyzed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines.

RESULTS

The primary search yielded 2538 articles. After screening and evaluation of eligibility, 117 studies were incorporated in the review. AI techniques were found to play a significant role in screening systems for pre-cancerous and cancerous cervical lesions. The accuracy of the algorithms in predicting cervical cancer varied from 70% to 100%. AI techniques make a distinction between cancerous and normal Pap smears with 80-100% accuracy. AI is expected to serve as a practical tool for doctors in making accurate clinical diagnoses. The reported sensitivity and specificity of AI in colposcopy for the detection of CIN2+ were 71.9-98.22% and 51.8-96.2%, respectively.

CONCLUSION

The present review highlights the acceptable performance of AI systems in the prediction, screening, or detection of cervical cancer and pre-cancerous lesions, especially when faced with a paucity of specialized centers or medical resources. In combination with human evaluation, AI could serve as a helpful tool in the interpretation of cervical smears or images.