Novel artificial intelligence machine learning approaches to precisely predict survival and site-specific recurrence in cervical cancer: A multi-institutional study

ConvXGB: A novel deep learning model to predict recurrence risk of early-stage cervical cancer following surgery using multiparametric MRI images

BACKGROUND

Accurate estimation of recurrence risk for cervical cancer plays a pivot role in making individualized treatment plans. We aimed to develop and externally validate an end-to-end deep learning model for predicting recurrence risk in cervical cancer patients following surgery by using multiparametric MRI images.

METHODS

The clinicopathologic data and multiparametric MRI images of 406 cervical cancer patients from three institutions were collected. We designed a novel deep learning model called "ConvXGB" for predicting recurrence risk by combining the convolutional neural network (CNN) and eXtreme Gradient Boost (XGBoost). The predictive performance of the ConvXGB model was evaluated using time-dependent area under curve (AUC), compared with the deep learning radio-clinical model, clinical model, conventional radiomics nomogram and an existing histology-specific tool. The potential of the ConvXGB model in predicting the recurrence-free survival (RFS) and overall survival (OS) was assessed.

RESULTS

The ConvXGB model outperformed other models in predicting recurrence risk, with AUCs for 1 and 3 year-RFS of 0.872(95% CI, 0.857-0.906) and 0.882(95% CI, 0.860-0.904) respectively in the test cohort. This model showed better discrimination, calibration and clinical utility. Grad-CAM analysis was adopted to help clinicians better understand the predictive results. Moreover, Kaplan-Meier survival analysis revealed that patients who were stratified into high-risk group by the ConvXGB model were significantly susceptible to higher cumulative recurrence risk rates and worse outcome.

CONCLUSION

The ConvXGB model allowed for predicting postoperative recurrence risk in cervical cancer patients and for stratifying the risk of RFS and OS.

Bidirectional recurrent neural network approach for predicting cervical cancer recurrence and survival

Cervical cancer is a deadly disease in women globally. There is a greater chance of getting rid of cervical cancer in case of earliest diagnosis. But for some patients, there is a chance of recurrence. The chances of treating the Recurrence of cervical carcinoma arelimited. The main objective of a research is to find the key features that will predict the cervical cancer recurrence and survival rates accurately by utilizing a neural network that is bidirectionally recurrent. The goal is to reduce risk factors of cervical cancer recurrence by identifying genes with positive coefficients and targeting them for preventive interventions. First step is identification of risk factors for cervical carcinoma recurrence by utilising clinical attributes. This research uses following Random forest, Logistic regression, Gradient boosting and support vector machine algorithms are applied for classification. Random forest offers the maximum precision of these four techniques at 91.2%. The second step is identifying long noncoding RNA (lnRNA) gene signatures among people with cervical carcinomaby implementingHSIC model. Intended to discover biomarkers in initial cervical carcinoma clinical data from people who experienced a distant repetition that could be connected to lnRNA gene signatures and utilized for forecasting survival rates using a bidirectional recurrent neural network(Bi-RNN). The results shows that Bi-RNN model effectively forecast the cervical cancer recurrence and survival.

Disease trend analysis platform accurately predicts the occurrence of cervical cancer under mixed diseases

Cervical cancer (CC) is one of the most common gynecological malignancies. Cytological screening, while being the most common and accurate method for detecting cervical cancer, is both time-consuming and costly. Predicting CC based on bioinformatics can assist in the rapid early screening of CC in clinical practice. Most recent CC prediction methods require a large amount of detection data or sequencing data and are not ideal for CC detection in complex disease samples. We developed the Disease trend analysis platform (Dtap), which can quickly predict the occurrence of diseases using only blood routine data. Blood routine data was collected from 1,292 cervical cancer patients, 4,860 patients with complex diseases, and 4,980 healthy individuals from various sources. The results show that the Dtap-based trend model maintained good and stable performance in the prediction task of multiple datasets as well as complex disease samples. Finally, we built DTAPCC (http://bioinfor.imu.edu.cn/dtapcc), a Dtap-based CC disease prediction platform, to help users quickly predict CC and visualize trend features.

Prognostic Factors for Cervical Cancer in Asian Populations: A Scoping Review of Research From 2013 to 2023

Cervical cancer is the fourth most common cancer among women worldwide, with particularly high incidence and mortality rates in low- and middle-income countries, with Asia reporting the highest number of cases in 2022. Despite this significant burden, the prognostic factors specific to Asian populations remain underexplored. This scoping review aimed to identify and evaluate prognostic factors associated with cervical cancer outcomes in Asia, focusing on clinical, socio-demographic, and treatment-related variables. The review followed the Arksey and O'Malley framework and included 44 studies published between 2013 and 2023. The majority of research was concentrated in East Asia, particularly in China, Japan, and South Korea. Key prognostic factors affecting overall survival and disease-free survival included tumor size, histology, age, lymphovascular invasion, and lymph node metastasis. Non-squamous cell carcinoma histology, especially adenocarcinoma, was consistently linked to poorer outcomes. Older age and medical comorbidities, such as anemia and diabetes, also negatively impacted survival. Treatment-related factors, though less frequently reported, demonstrated the significance of adjuvant therapy, chemotherapy, and treatment intensity in improving outcomes. This review underscores the complexity of cervical cancer prognosis in Asian populations and highlights the need for targeted research and region-specific interventions to address the rising incidence of cervical cancer. It also highlights the scarcity of research on cervical cancer prognostic factors in West, Central, and South Asian countries. Future research should aim to address the gaps in understanding treatment-related factors and explore the potential for region-specific interventions to improve outcomes in cervical cancer across Asia.

Integrating cat boost algorithm with triangulating feature importance to predict survival outcome in recurrent cervical cancer

Cervical cancer is one of the most dangerous malignancies in women. Prolonged survival times are made possible by breakthroughs in early recognition and efficient treatment of a disease.The existing methods are lagging on finding the important attributes to predict the survival outcome. The main objective of this study is to find individuals with cervical cancer who are at greater risk of death from recurrence by predicting the survival.A novel approach in a proposed technique is Triangulating feature importance to find the important risk factors through which the treatment may vary to improve the survival outcome.Five algorithms Support vector machine, Naive Bayes, supervised logistic regression, decision tree algorithm, Gradient boosting, and random forest are used to build the concept. Conventional attribute selection methods like information gain (IG), FCBF, and ReliefFare employed. The recommended classifier is evaluated for Precision, Recall, F1, Mathews Correlation Coefficient (MCC), Classification Accuracy (CA), and Area under curve (AUC) using various methods. Gradient boosting algorithm (CAT BOOST) attains the highest accuracy value of 0.99 to predict survival outcome of recurrence cervical cancer patients. The proposed outcome of the research is to identify the important risk factors through which the survival outcome of the patients improved.

Artificial Intelligence Needs Data: Challenges Accessing Italian Databases to Train AI

Population biobanks are an increasingly important infrastructure to support research and will be a much-needed resource in the delivery of personalised medicine. Artificial intelligence (AI) systems can process and cross-link very large amounts of data quickly and be used not only for improving research power but also for helping with complex diagnosis and prediction of diseases based on health profiles. AI, therefore, potentially has a critical role to play in personalised medicine, and biobanks can provide a lot of the necessary baseline data related to healthy populations that will enable the development of AI tools. To develop these tools, access to personal data, and in particular, sensitive data, is required. Such data could be accessed from biobanks. Biobanks are a valuable resource for research but accessing and using the data contained within such biobanks raise a host of legal, ethical, and social issues (ELSI). This includes the appropriate consent to manage the collection, storage, use, and sharing of samples and data, and appropriate governance models that provide oversight of secondary use of samples and data. Biobanks have developed new consent models and governance tools to enable access that address some of these ELSI-related issues. In this paper, we consider whether such governance frameworks can enable access to biobank data to develop AI. As Italy has one of the most restrictive regulatory frameworks on the use of genetic data in Europe, we examine the regulatory framework in Italy. We also look at the proposed changes under the European Health Data Space (EHDS). We conclude by arguing that currently, regulatory frameworks are misaligned and unless addressed, accessing data within Italian biobanks to train AI will be severely limited.

A novel higher performance nomogram based on explainable machine learning for predicting mortality risk in stroke patients within 30 days based on clinical features on the first day ICU admission

BACKGROUND

This study aimed to develop a higher performance nomogram based on explainable machine learning methods, and to predict the risk of death of stroke patients within 30 days based on clinical characteristics on the first day of intensive care units (ICU) admission.

METHODS

Data relating to stroke patients were extracted from the Medical Information Marketplace of the Intensive Care (MIMIC) IV and III database. The LightGBM machine learning approach together with Shapely additive explanations (termed as explain machine learning, EML) was used to select clinical features and define cut-off points for the selected features. These selected features and cut-off points were then evaluated using the Cox proportional hazards regression model and Kaplan-Meier survival curves. Finally, logistic regression-based nomograms for predicting 30-day mortality of stroke patients were constructed using original variables and variables dichotomized by cut-off points, respectively. The performance of two nomograms were evaluated in overall and individual dimension.

RESULTS

A total of 2982 stroke patients and 64 clinical features were included, and the 30-day mortality rate was 23.6% in the MIMIC-IV datasets. 10 variables ("sofa (sepsis-related organ failure assessment)", "minimum glucose", "maximum sodium", "age", "mean spo2 (blood oxygen saturation)", "maximum temperature", "maximum heart rate", "minimum bun (blood urea nitrogen)", "minimum wbc (white blood cells)" and "charlson comorbidity index") and respective cut-off points were defined from the EML. In the Cox proportional hazards regression model (Cox regression) and Kaplan-Meier survival curves, after grouping stroke patients according to the cut-off point of each variable, patients belonging to the high-risk subgroup were associated with higher 30-day mortality than those in the low-risk subgroup. The evaluation of nomograms found that the EML-based nomogram not only outperformed the conventional nomogram in NIR (net reclassification index), brier score and clinical net benefits in overall dimension, but also significant improved in individual dimension especially for low "maximum temperature" patients.

CONCLUSIONS

The 10 selected first-day ICU admission clinical features require greater attention for stroke patients. And the nomogram based on explainable machine learning will have greater clinical application.

Pre-treatment T2-weighted magnetic resonance radiomics for prediction of loco-regional recurrence after image-guided adaptive brachytherapy for locally advanced cervical cancer

Purpose

The aim of this study was to investigate the predictive value of radiomic features of pre-treatment T2-weighted magnetic resonance images (MRI) for clinical outcomes of radiotherapy in cervical cancer patients.

Material and methods

Ninety cervical cancer patients with stage IB-IVA were retrospectively analyzed. All patients received definitive radiotherapy with or without concurrent chemotherapy. Radiomic features were extracted from gross tumor volume (GTV) on pre-treatment T2-weighted MRI. The association between radiomic features and loco-regional recurrence (LRR) was analyzed with Student's t test, and false discovery rate was controlled using Storey method. Multivariate analysis with significant radiomic features with p-value < 0.01 and known clinical prognostic factors was performed using Cox proportional hazard model.

Results

The majority of patients were stage IIIB (47.8%) and stage IIB (36.7%), and the most common histology was squamous cell carcinoma (74.5%). The median GTV volume was 37.5 ml (IQR, 16.3-93.1). The median dose of D90 received by high-risk clinical target volume (HR-CTV) was 86.2 Gy (IQR, 67.2-94.2). In a median follow-up time of 29.2 months, 12 of the 90 patients (13.3%) developed LRR. Eighty radiomic features were collected. There were four radiomic features, which showed significant correlation with LRR: Maximum intensity (p = 0.0002), Correlation135 GLCM (p = 0.0014), Correlation90 (p = 0.0015), and Correlation45 (p = 0.0034). Cox regression analysis yielded a significant hazard ratio for the maximum intensity (p = 0.038) and Correlation135 GLCM (p = 0.013) features. There was no statistically significant association for overall survival with any radiomic features.

Conclusions

The maximum intensity and Correlation135 GLCM radiomic features of the pre-treatment T2-weighted MR images are predictive of loco-regional recurrence in cervical cancer patients after definitive radiotherapy with 3D-IGABT.

A Deep Learning-Based Approach for Cervical Cancer Classification Using 3D CNN and Vision Transformer

Cervical cancer is a significant health problem worldwide, and early detection and treatment are critical to improving patient outcomes. To address this challenge, a deep learning (DL)-based cervical classification system is proposed using 3D convolutional neural network and Vision Transformer (ViT) module. The proposed model leverages the capability of 3D CNN to extract spatiotemporal features from cervical images and employs the ViT model to capture and learn complex feature representations. The model consists of an input layer that receives cervical images, followed by a 3D convolution block, which extracts features from the images. The feature maps generated are down-sampled using max-pooling block to eliminate redundant information and preserve important features. Four Vision Transformer models are employed to extract efficient feature maps of different levels of abstraction. The output of each Vision Transformer model is an efficient set of feature maps that captures spatiotemporal information at a specific level of abstraction. The feature maps generated by the Vision Transformer models are then supplied into the 3D feature pyramid network (FPN) module for feature concatenation. The 3D squeeze-and-excitation (SE) block is employed to obtain efficient feature maps that recalibrate the feature responses of the network based on the interdependencies between different feature maps, thereby improving the discriminative power of the model. At last, dimension minimization of feature maps is executed using 3D average pooling layer. Its output is then fed into a kernel extreme learning machine (KELM) for classification into one of the five classes. The KELM uses radial basis kernel function (RBF) for mapping features in high-dimensional feature space and classifying the input samples. The superiority of the proposed model is known using simulation results, achieving an accuracy of 98.6%, demonstrating its potential as an effective tool for cervical cancer classification. Also, it can be used as a diagnostic supportive tool to assist medical experts in accurately identifying cervical cancer in patients.

Prediction of Cervical Cancer Patients' Survival Period with Machine Learning Techniques

OBJECTIVES

The objective of this research is to apply machine learning (ML) algorithms to predict the survival of cervical cancer patients. The aim was to address the limitations of traditional statistical methods, which often fail to provide accurate answers due to the complexity of the problem.

METHODS

This research employed visualization techniques for initial data understanding. Subsequently, ML algorithms were used to develop both classification and regression models for survival prediction. In the classification models, we trained the algorithms to predict the time interval between the initial diagnosis and the patient's death. The intervals were categorized as "<6 months," "6 months to 3 years," "3 years to 5 years," and ">5 years." The regression model aimed to predict survival time (in months). We used attribute weights to gain insights into the model, highlighting features with a significant impact on predictions and offering valuable insights into the model's behavior and decision-making process.

RESULTS

The gradient boosting trees algorithm achieved an 81.55% accuracy in the classification model, while the random forest algorithm excelled in the regression model, with a root mean square error of 22.432. Notably, radiation doses around the affected areas significantly influenced survival duration.

CONCLUSIONS

Machine learning demonstrated the ability to provide high-accuracy predictions of survival periods in both classification and regression problems. This suggests its potential use as a decision-support tool in the process of treatment planning and resource allocation for each patient.

Cervical cancer survival prediction by machine learning algorithms: a systematic review

BACKGROUND

Cervical cancer is a common malignant tumor of the female reproductive system and is considered a leading cause of mortality in women worldwide. The analysis of time to event, which is crucial for any clinical research, can be well done with the method of survival prediction. This study aims to systematically investigate the use of machine learning to predict survival in patients with cervical cancer.

METHOD

An electronic search of the PubMed, Scopus, and Web of Science databases was performed on October 1, 2022. All articles extracted from the databases were collected in an Excel file and duplicate articles were removed. The articles were screened twice based on the title and the abstract and checked again with the inclusion and exclusion criteria. The main inclusion criterion was machine learning algorithms for predicting cervical cancer survival. The information extracted from the articles included authors, publication year, dataset details, survival type, evaluation criteria, machine learning models, and the algorithm execution method.

RESULTS

A total of 13 articles were included in this study, most of which were published from 2018 onwards. The most common machine learning models were random forest (6 articles, 46%), logistic regression (4 articles, 30%), support vector machines (3 articles, 23%), ensemble and hybrid learning (3 articles, 23%), and Deep Learning (3 articles, 23%). The number of sample datasets in the study varied between 85 and 14946 patients, and the models were internally validated except for two articles. The area under the curve (AUC) range for overall survival (0.40 to 0.99), disease-free survival (0.56 to 0.88), and progression-free survival (0.67 to 0.81), respectively from (lowest to highest) received. Finally, 15 variables with an effective role in predicting cervical cancer survival were identified.

CONCLUSION

Combining heterogeneous multidimensional data with machine learning techniques can play a very influential role in predicting cervical cancer survival. Despite the benefits of machine learning, the problem of interpretability, explainability, and imbalanced datasets is still one of the biggest challenges. Providing machine learning algorithms for survival prediction as a standard requires further studies.

Gynecological cancer prognosis using machine learning techniques: A systematic review of last three decades (1990–2022)

OBJECTIVE

Many Computer Aided Prognostic (CAP) systems based on machine learning techniques have been proposed in the field of oncology. The objective of this systematic review was to assess and critically appraise the methodologies and approaches used in predicting the prognosis of gynecological cancers using CAPs.

METHODS

Electronic databases were used to systematically search for studies utilizing machine learning methods in gynecological cancers. Study risk of bias (ROB) and applicability were assessed using the PROBAST tool. 139 studies met the inclusion criteria, of which 71 predicted outcomes for ovarian cancer patients, 41 predicted outcomes for cervical cancer patients, 28 predicted outcomes for uterine cancer patients, and 2 predicted outcomes for gynecological malignancies broadly.

RESULTS

Random forest (22.30 %) and support vector machine (21.58 %) classifiers were used most commonly. Use of clinicopathological, genomic and radiomic data as predictors was observed in 48.20 %, 51.08 % and 17.27 % of studies, respectively, with some studies using multiple modalities. 21.58 % of studies were externally validated. Twenty-three individual studies compared ML and non-ML methods. Study quality was highly variable and methodologies, statistical reporting and outcome measures were inconsistent, preventing generalized commentary or meta-analysis of performance outcomes.

CONCLUSION

There is significant variability in model development when prognosticating gynecological malignancies with respect to variable selection, machine learning (ML) methods and endpoint selection. This heterogeneity prevents meta-analysis and conclusions regarding the superiority of ML methods. Furthermore, PROBAST-mediated ROB and applicability analysis demonstrates concern for the translatability of existing models. This review identifies ways that this can be improved upon in future works to develop robust, clinically translatable models within this promising field.

A 5‐year survival status prognosis of nonmetastatic cervical cancer patients through machine learning algorithms

BACKGROUND

Prediction models with high accuracy rates for nonmetastatic cervical cancer (CC) patients are limited. This study aimed to construct and compare predictive models on the basis of machine learning (ML) algorithms for predicting the 5-year survival status of CC patients through using the Surveillance, Epidemiology, and End Results public database of the National Cancer Institute.

METHODS

The data registered from 2004 to 2016 were extracted and randomly divided into training and validation cohorts (8:2). The least absolute shrinkage and selection operator (LASSO) regression was employed to identify significant factors. Then, four predictive models were constructed, including logistic regression (LR), random forest (RF), support vector machine (SVM), and extreme gradient boosting (XGBoost). The predictive models were evaluated and compared using Receiver-operating characteristics with areas under the curves (AUCs) and decision curve analysis (DCA), respectively.

RESULTS

A total of 13,802 patients were involved and classified into training (N = 11,041) and validation (N = 2761) cohorts. By using the LASSO regression method, seven factors were identified. In the training cohort, the XGBoost model showed the best performance (AUC = 0.8400) compared to the other three models (all p < 0.05 by Delong's test). In the validation cohort, the XGBoost model also demonstrated a superior prediction ability (AUC = 0.8365) than LR and SVM models (both p < 0.05 by Delong's test), although the difference was not statistically significant between the XGBoost and the RF models (p = 0.4251 by Delong's test). Based on the DCA results, the XGBoost model was also superior, and feature importance analysis indicated that the tumor stage was the most important variable among the seven factors.

CONCLUSIONS

The XGBoost model proved to be an effective algorithm with better prediction abilities. This model is proposed to support better decision-making for nonmetastatic CC patients in the future.

Hybrid Model for Detection of Cervical Cancer Using Causal Analysis and Machine Learning Techniques

Cervical cancer has become the third most common form of cancer in the in-universe, after the widespread breast cancer. Human papillomavirus risk of infection is linked to the majority of cancer cases. Preventive care, the most expensive way of fighting cancer, can protect about 37% of cancer cases. The Pap smear examination is a standard screening procedure for the initial screening of cervical cancer. However, this manual test procedure generates many false-positive outcomes due to individual errors. Various researchers have extensively investigated machine learning (ML) methods for classifying cervical Pap cells to enhance manual testing. The random forest method is the most popular method for anticipating features from a high-dimensional cancer image dataset. However, the random forest method can get too slow and inefficient for real-time forecasts when too many decision trees are used. This research proposed an efficient feature selection and prediction model for cervical cancer datasets using Boruta analysis and SVM method to deal with this challenge. A Boruta analysis method is used. It is improved from of random forest method and mainly discovers feature subsets from the data source that are significant to assigned classification activity. The proposed model's primary aim is to determine the importance of cervical cancer screening factors for classifying high-risk patients depending on the findings. This research work analyses cervical cancer and various risk factors to help detect cervical cancer. The proposed model Boruta with SVM and various popular ML models are implemented using Python and various performance measuring parameters, i.e., accuracy, precision, F1–Score, and recall. However, the proposed Boruta analysis with SVM performs outstanding over existing methods.

An Integrated Approach for Cancer Survival Prediction Using Data Mining Techniques

Ovarian cancer is the third most common gynecologic cancers worldwide. Advanced ovarian cancer patients bear a significant mortality rate. Survival estimation is essential for clinicians and patients to understand better and tolerate future outcomes. The present study intends to investigate different survival predictors available for cancer prognosis using data mining techniques. Dataset of 140 advanced ovarian cancer patients containing data from different data profiles (clinical, treatment, and overall life quality) has been collected and used to foresee cancer patients' survival. Attributes from each data profile have been processed accordingly. Clinical data has been prepared corresponding to missing values and outliers. Treatment data including varying time periods were created using sequence mining techniques to identify the treatments given to the patients. And lastly, different comorbidities were combined into a single factor by computing Charlson Comorbidity Index for each patient. After appropriate preprocessing, the integrated dataset is classified using appropriate machine learning algorithms. The proposed integrated model approach gave the highest accuracy of 76.4% using ensemble technique with sequential pattern mining including time intervals of 2 months between treatments. Thus, the treatment sequences and, most importantly, life quality attributes significantly contribute to the survival prediction of cancer patients.