Comparison of time-to-event machine learning models in predicting oral cavity cancer prognosis

DeepOmicsSurv: a deep learning-based model for survival prediction of oral cancer

OBJECTIVE

Oral cancer is an important health challenge worldwide and accurate survival time prediction of this disease can guide treatment decisions. This study aims to propose a deep learning-based model, DeepOmicsSurv, to predict survival in oral cancer patients using clinical and multi-omics data.

METHODS

DeepOmicsSurv builds on the DeepSurv model, incorporating multi-head attention convolutional layers, dropout, pooling, and batch normalization to boost its strength and precision. Various dimensionality reduction techniques, including Principal Component Analysis (PCA), Kernel PCA, Non-Negative Matrix Factorization (NMF), Singular Value Decomposition (SVD), Partial Least Squares (PLS), Multidimensional Scaling (MDS), and Autoencoders, were employed to manage the high-dimensional omics data. The model's performance was evaluated against DeepSurv, DeepHit, Cox Proportional Hazards (CoxPH), Convolutional Neural Networks (CNN), and Recurrent Neural Networks (RNN). Additionally, SHapley Additive Explanations (SHAP) was used to analyze the impact of clinical features on survival predictions.

RESULTS

DeepOmicsSurv achieved a C-index of 0.966, MSE of 0.0138, RMSE of 0.1174, MAE of 0.0795, and MedAE of 0.0515, outperforming other deep learning models. Among various dimensionality reduction techniques, autoencoder performed the best with DeepOmicsSurv. SHAP analysis showed that Age, AJCC N Stage, alcohol history and patient smoking history are prevalent clinical features for survival time.

CONCLUSION

In conclusion, DeepOmicsSurv has the potential to predict survival time in oral cancer patients. This model achieved high accuracy with various data types including Clinical, DNAmethylation + clinical, mRNA + clinical, Copy number alteration + clinical, or multi-omics data. Additionally, SHAP analysis reveals clinical factors that influence survival time.

Prediction of Medication-Related Osteonecrosis of the Jaw in Patients Receiving Antiresorptive Therapy Using Machine Learning Models

BACKGROUND

Medication-related osteonecrosis of the jaw (MRONJ) is a serious complication associated with the use of antiresorptive agents, impacting patient quality of life and treatment outcomes. Predictive modeling may aid in a better understanding of MRONJ development.

PURPOSE

The study aimed to evaluate machine learning (ML)-based models for predicting MRONJ in patients receiving antiresorptive therapy.

STUDY DESIGN, SETTING, SAMPLE

This retrospective in silico study analyzed electronic medical records from Thammasat University Hospital, covering the period from January 2012 to December 2022. The sample included subjects receiving antiresorptive therapy, excluding those with a history of radiation therapy or metastatic jaw disease.

PREDICTOR VARIABLES

The primary predictor variable was the predicted probability of MRONJ development from the ML models.

OUTCOME VARIABLES

The outcome variable was MRONJ status coded as present or absent based on chart review.

COVARIATES

Covariates included demographic data, MRONJ occurrence, location and staging of MRONJ, comorbidities, diseases related to antiresorptive agents, types of antiresorptive agents, therapy duration, concurrent medications, blood calcium levels, and dental factors.

ANALYSES

Model performance was assessed via accuracy, sensitivity, specificity, positive and negative predictive values, and the area under the receiver operating characteristic curve. Additionally, univariate and multivariate Cox regression analyses were conducted to identify factors significantly associated with MRONJ development. P ≤ .05 was statistically significant.

RESULTS

The study analyzed data from 5,305 subjects with a mean age of 75 ± 11.1 years, predominantly female. MRONJ was observed in 81 cases (1.5%), with a median time to development of 33 months (interquartile range = 3). Among the 6 models tested, the best-performing model had an accuracy of 0.95 and an area under the receiver operating characteristic curve of 0.89-0.90. Significant predictors identified through Cox regression included metabolic syndrome (hazard ratio = 14.064, 95% confidence interval = 1.111-178.067, P = .041) and patients receiving intravenous pamidronate (hazard ratio = 5.932, 95% confidence interval = 1.755-20.051, P = .004), indicating their association with MRONJ development.

CONCLUSIONS AND RELEVANCE

ML-based predictive and time-to-event models effectively predict MRONJ risk, aiding in the strategic prevention and management for patients undergoing antiresorptive therapy.

Integrating deep learning in public health: a novel approach to PICC-RVT risk assessment

Background

Machine learning is pivotal for predicting Peripherally Inserted Central Catheter-related venous thrombosis (PICC-RVT) risk, facilitating early diagnosis and proactive treatment. Existing models often assess PICC-RVT risk as static and discrete outcomes, which may limit their practical application.

Objectives

This study aims to evaluate the effectiveness of seven diverse machine learning algorithms, including three deep learning and four traditional machine learning models, that incorporate time-series data to assess PICC-RVT risk. It also seeks to identify key predictive factors for PICC-RVT using these models.

Methods

We conducted a retrospective multi-center cohort study involving 5,272 patients who underwent PICC placement. After preprocessing patient data, the models were trained. Demographic, clinical pathology, and treatment data were analyzed to identify predictive factors. A variable analysis was then conducted to determine the most significant predictors of PICC-RVT. Model performance was evaluated using the Concordance Index (c-index) and the composite Brier score, and the Intraclass Correlation Coefficient (ICC) from cross-validation folds assessed model stability.

Results

Deep learning models generally outperformed traditional machine learning models in terms of predictive accuracy (mean c-index: 0.949 vs. 0.732; mean integrated Brier score: 0.046 vs. 0.093). Specifically, the DeepSurv model demonstrated exceptional precision in risk assessment (c-index: 0.95). Stability varied with the number of predictive factors, with Cox-Time showing the highest ICC (0.974) with 16 predictive factors, and DeepSurv the most stable with 26 predictive factors (ICC: 0.983). Key predictors across models included albumin levels, prefill sealant type, and activated partial thromboplastin time.

Conclusion

Machine learning models that incorporate time-to-event data can effectively predict PICC-RVT risk. The DeepSurv model, in particular, shows excellent discriminative and calibration capabilities. Albumin levels, type of prefill sealant, and activated partial thromboplastin time are critical indicators for identifying and managing high-risk PICC-RVT patients.

Prognosing post-treatment outcomes of head and neck cancer using structured data and machine learning: A systematic review

BACKGROUND

This systematic review aimed to evaluate the performance of machine learning (ML) models in predicting post-treatment survival and disease progression outcomes, including recurrence and metastasis, in head and neck cancer (HNC) using clinicopathological structured data.

METHODS

A systematic search was conducted across the Medline, Scopus, Embase, Web of Science, and Google Scholar databases. The methodological characteristics and performance metrics of studies that developed and validated ML models were assessed. The risk of bias was evaluated using the Prediction model Risk Of Bias ASsessment Tool (PROBAST).

RESULTS

Out of 5,560 unique records, 34 articles were included. For survival outcome, the ML model outperformed the Cox proportional hazards model in time-to-event analyses for HNC, with a concordance index of 0.70-0.79 vs. 0.66-0.76, and for all sub-sites including oral cavity (0.73-0.89 vs. 0.69-0.77) and larynx (0.71-0.85 vs. 0.57-0.74). In binary classification analysis, the area under the receiver operating characteristics (AUROC) of ML models ranged from 0.75-0.97, with an F1-score of 0.65-0.89 for HNC; AUROC of 0.61-0.91 and F1-score of 0.58-0.86 for the oral cavity; and AUROC of 0.76-0.97 and F1-score of 0.63-0.92 for the larynx. Disease-specific survival outcomes showed higher performance than overall survival outcomes, but the performance of ML models did not differ between three- and five-year follow-up durations. For disease progression outcomes, no time-to-event metrics were reported for ML models. For binary classification of the oral cavity, the only evaluated subsite, the AUROC ranged from 0.67 to 0.97, with F1-scores between 0.53 and 0.89.

CONCLUSIONS

ML models have demonstrated considerable potential in predicting post-treatment survival and disease progression, consistently outperforming traditional linear models and their derived nomograms. Future research should incorporate more comprehensive treatment features, emphasize disease progression outcomes, and establish model generalizability through external validations and the use of multicenter datasets.

Development and Validation of Prognostic Models Using Radiomic Features from Pre-Treatment Positron Emission Tomography (PET) Images in Head and Neck Squamous Cell Carcinoma (HNSCC) Patients

High-dimensional radiomics features derived from pre-treatment positron emission tomography (PET) images offer prognostic insights for patients with head and neck squamous cell carcinoma (HNSCC). Using 124 PET radiomics features and clinical variables (age, sex, stage of cancer, site of cancer) from a cohort of 232 patients, we evaluated four survival models-penalized Cox model, random forest, gradient boosted model and support vector machine-to predict all-cause mortality (ACM), locoregional recurrence/residual disease (LR) and distant metastasis (DM) probability during 36, 24 and 24 months of follow-up, respectively. We developed models with five-fold cross-validation, selected the best-performing model for each outcome based on the concordance index (C-statistic) and the integrated Brier score (IBS) and validated them in an independent cohort of 102 patients. The penalized Cox model demonstrated better performance for ACM (C-statistic = 0.70, IBS = 0.12) and DM (C-statistic = 0.70, IBS = 0.08) while the random forest model displayed better performance for LR (C-statistic = 0.76, IBS = 0.07). We conclude that the ML-based prognostic model can aid clinicians in quantifying prognosis and determining effective treatment strategies, thereby improving favorable outcomes in HNSCC patients.

Survival estimation of oral cancer using fuzzy deep learning

BACKGROUND

Oral cancer is a deadly disease and a major cause of morbidity and mortality worldwide. The purpose of this study was to develop a fuzzy deep learning (FDL)-based model to estimate the survival time based on clinicopathologic data of oral cancer.

METHODS

Electronic medical records of 581 oral squamous cell carcinoma (OSCC) patients, treated with surgery with or without radiochemotherapy, were collected retrospectively from the Oral and Maxillofacial Surgery Clinic and the Regional Cancer Center from 2011 to 2019. The deep learning (DL) model was trained to classify survival time classes based on clinicopathologic data. Fuzzy logic was integrated into the DL model and trained to create FDL-based models to estimate the survival time classes.

RESULTS

The performance of the models was evaluated on a test dataset. The performance of the DL and FDL models for estimation of survival time achieved an accuracy of 0.74 and 0.97 and an area under the receiver operating characteristic (AUC) curve of 0.84 to 1.00 and 1.00, respectively.

CONCLUSIONS

The integration of fuzzy logic into DL models could improve the accuracy to estimate survival time based on clinicopathologic data of oral cancer.

Leveraging artificial intelligence for perioperative cancer risk assessment of oral potentially malignant disorders

Oral potentially malignant disorders (OPMDs) are mucosal conditions with an inherent disposition to develop oral squamous cell carcinoma. Surgical management is the most preferred strategy to prevent malignant transformation in OPMDs, and surgical approaches to treatment include conventional scalpel excision, laser surgery, cryotherapy, and photodynamic therapy. However, in reality, since all patients with OPMDs will not develop oral squamous cell carcinoma in their lifetime, there is a need to stratify patients according to their risk of malignant transformation to streamline surgical intervention for patients with the highest risks. Artificial intelligence (AI) has the potential to integrate disparate factors influencing malignant transformation for robust, precise, and personalized cancer risk stratification of OPMD patients than current methods to determine the need for surgical resection, excision, or re-excision. Therefore, this article overviews existing AI models and tools, presents a clinical implementation pathway, and discusses necessary refinements to aid the clinical application of AI-based platforms for cancer risk stratification of OPMDs in surgical practice.

Deep learning in oral cancer- a systematic review

BACKGROUND

Oral cancer is a life-threatening malignancy, which affects the survival rate and quality of life of patients. The aim of this systematic review was to review deep learning (DL) studies in the diagnosis and prognostic prediction of oral cancer.

METHODS

This systematic review was conducted following the PRISMA guidelines. Databases (Medline via PubMed, Google Scholar, Scopus) were searched for relevant studies, from January 2000 to June 2023.

RESULTS

Fifty-four qualified for inclusion, including diagnostic (n = 51), and prognostic prediction (n = 3). Thirteen studies showed a low risk of biases in all domains, and 40 studies low risk for concerns regarding applicability. The performance of DL models was reported of the accuracy of 85.0-100%, F1-score of 79.31 - 89.0%, Dice coefficient index of 76.0 - 96.3% and Concordance index of 0.78-0.95 for classification, object detection, segmentation, and prognostic prediction, respectively. The pooled diagnostic odds ratios were 2549.08 (95% CI 410.77-4687.39) for classification studies.

CONCLUSIONS

The number of DL studies in oral cancer is increasing, with a diverse type of architectures. The reported accuracy showed promising DL performance in studies of oral cancer and appeared to have potential utility in improving informed clinical decision-making of oral cancer.

Development of interpretable machine learning models associated with environmental chemicals to predict all-cause and specific-cause mortality：A longitudinal study based on NHANES

Limited information is available on potential predictive value of environmental chemicals for mortality. Our study aimed to investigate the associations between 43 of 8 classes representative environmental chemicals in serum/urine and mortality, and further develop the interpretable machine learning models associated with environmental chemicals to predict mortality. A total of 1602 participants were included from the National Health and Nutrition Examination Survey (NHANES). During 154,646 person-months of follow-up, 127 deaths occurred. We found that machine learning showed promise in predicting mortality. CoxPH was selected as the optimal model for predicting all-cause mortality with time-dependent AUROC of 0.953 (95%CI: 0.951-0.955). Coxnet was the best model for predicting cardiovascular disease (CVD) and cancer mortality with time-dependent AUROCs of 0.935 (95%CI: 0.933-0.936) and 0.850 (95%CI: 0.844-0.857). Based on clinical variables, adding environmental chemicals could enhance the predictive ability of cancer mortality (P < 0.05). Some environmental chemicals contributed more to the models than traditional clinical variables. Combined the results of association and prediction models by interpretable machine learning analyses, we found urinary methyl paraben (MP) and urinary 2-napthol (2-NAP) were negatively associated with all-cause mortality, while serum cadmium (Cd) was positively associated with all-cause mortality. Urinary bisphenol A (BPA) was positively associated with CVD mortality.

Anoikis-related mRNA-lncRNA and DNA methylation profiles for overall survival prediction in breast cancer patients

As a type of programmed cell death, anoikis resistance plays an essential role in tumor metastasis, allowing cancer cells to survive in the systemic circulation and as a key pathway for regulating critical biological processes. We conducted an exploratory analysis to improve risk stratification and optimize adjuvant treatment choices for patients with breast cancer, and identify multigene features in mRNA and lncRNA transcriptome profiles associated with anoikis. First, the variance selection method filters low information content genes in RNA sequence and then extracts the mRNA and lncRNA expression data base on annotation files. Then, the top ten key mRNAs are screened out through the PPI network. Pearson analysis has been employed to identify lncRNAs related to anoikis, and the prognosis-related lncRNAs are selected using Univariate Cox regression and machine learning. Finally, we identified a group of RNAs (including ten mRNAs and six lncRNAs) and integrated the expression data of 16 genes to construct a risk-scoring system for BRCA prognosis and drug sensitivity analysis. The risk score's validity has been evaluated with the ROC curve, Kaplan-Meier survival curve analysis and decision curve analysis (DCA). For the methylation data, we have obtained 169 anoikis-related prognostic methylation sites, integrated these sites with 16 RNA features and further used the deep learning model to evaluate and predict the survival risk of patients. The developed anoikis feature is demonstrated a consistency index (C-index) of 0.778, indicating its potential to predict the survival probability of breast cancer patients using deep learning methods.

Brain health scores to predict neurological outcomes from electronic health records

BACKGROUND

Preserving brain health is a critical priority in primary care, yet screening for these risk factors in face-to-face primary care visits is challenging to scale to large populations. We aimed to develop automated brain health risk scores calculated from data in the electronic health record (EHR) enabling population-wide brain health screening in advance of patient care visits.

METHODS

This retrospective cohort study included patients with visits to an outpatient neurology clinic at Massachusetts General Hospital, between January 2010 and March 2021. Survival analysis with an 11-year follow-up period was performed to predict the risk of intracranial hemorrhage, ischemic stroke, depression, death and composite outcome of dementia, Alzheimer's disease, and mild cognitive impairment. Variables included age, sex, vital signs, laboratory values, employment status and social covariates pertaining to marital, tobacco and alcohol status. Random sampling was performed to create a training (70%) set for hyperparameter tuning in internal 5-fold cross validation and an external hold-out testing (30%) set of patients, both stratified by age. Risk ratios for high and low risk groups were evaluated in the hold-out test set, using 1000 bootstrapping iterations to calculate 95% confidence intervals (CI).

RESULTS

The cohort comprised 17,040 patients with an average age of 49 ± 15.6 years; majority were males (57 %), White (78 %) and non-Hispanic (80 %). The low and high groups average risk ratios [95 % CI] were: intracranial hemorrhage 0.46 [0.45-0.48] and 2.07 [1.95-2.20], ischemic stroke 0.57 [0.57-0.59] and 1.64 [1.52-1.69], depression 0.68 [0.39-0.74] and 1.29 [0.78-1.38], composite of dementia 0.27 [0.26-0.28] and 3.52 [3.18-3.81] and death 0.24 [0.24-0.24] and 3.96 [3.91-4.00].

CONCLUSIONS

Simple risk scores derived from routinely collected EHR accurately quantify the risk of developing common neurologic and psychiatric diseases. These scores can be computed automatically, prior to medical care visits, and may thus be useful for large-scale brain health screening.

Application of machine learning in predicting survival outcomes involving real-world data: a scoping review

BACKGROUND

Despite the interest in machine learning (ML) algorithms for analyzing real-world data (RWD) in healthcare, the use of ML in predicting time-to-event data, a common scenario in clinical practice, is less explored. ML models are capable of algorithmically learning from large, complex datasets and can offer advantages in predicting time-to-event data. We reviewed the recent applications of ML for survival analysis using RWD in healthcare.

METHODS

PUBMED and EMBASE were searched from database inception through March 2023 to identify peer-reviewed English-language studies of ML models for predicting time-to-event outcomes using the RWD. Two reviewers extracted information on the data source, patient population, survival outcome, ML algorithms, and the Area Under the Curve (AUC).

RESULTS

Of 257 citations, 28 publications were included. Random survival forests (N = 16, 57%) and neural networks (N = 11, 39%) were the most popular ML algorithms. There was variability across AUC for these ML models (median 0.789, range 0.6-0.950). ML algorithms were predominately considered for predicting overall survival in oncology (N = 12, 43%). ML survival models were often used to predict disease prognosis or clinical events (N = 27, 96%) in the oncology, while less were used for treatment outcomes (N = 1, 4%).

CONCLUSIONS

The ML algorithms, random survival forests and neural networks, are mainly used for RWD to predict survival outcomes such as disease prognosis or clinical events in the oncology. This review shows that more opportunities remain to apply these ML algorithms to inform treatment decision-making in clinical practice. More methodological work is also needed to ensure the utility and applicability of ML models in survival outcomes.

Predicting Colorectal Cancer Survival Using Time-to-Event Machine Learning: Retrospective Cohort Study

BACKGROUND

Machine learning (ML) methods have shown great potential in predicting colorectal cancer (CRC) survival. However, the ML models introduced thus far have mainly focused on binary outcomes and have not considered the time-to-event nature of this type of modeling.

OBJECTIVE

This study aims to evaluate the performance of ML approaches for modeling time-to-event survival data and develop transparent models for predicting CRC-specific survival.

METHODS

The data set used in this retrospective cohort study contains information on patients who were newly diagnosed with CRC between December 28, 2012, and December 27, 2019, at West China Hospital, Sichuan University. We assessed the performance of 6 representative ML models, including random survival forest (RSF), gradient boosting machine (GBM), DeepSurv, DeepHit, neural net-extended time-dependent Cox (or Cox-Time), and neural multitask logistic regression (N-MTLR) in predicting CRC-specific survival. Multiple imputation by chained equations method was applied to handle missing values in variables. Multivariable analysis and clinical experience were used to select significant features associated with CRC survival. Model performance was evaluated in stratified 5-fold cross-validation repeated 5 times by using the time-dependent concordance index, integrated Brier score, calibration curves, and decision curves. The SHapley Additive exPlanations method was applied to calculate feature importance.

RESULTS

A total of 2157 patients with CRC were included in this study. Among the 6 time-to-event ML models, the DeepHit model exhibited the best discriminative ability (time-dependent concordance index 0.789, 95% CI 0.779-0.799) and the RSF model produced better-calibrated survival estimates (integrated Brier score 0.096, 95% CI 0.094-0.099), but these are not statistically significant. Additionally, the RSF, GBM, DeepSurv, Cox-Time, and N-MTLR models have comparable predictive accuracy to the Cox Proportional Hazards model in terms of discrimination and calibration. The calibration curves showed that all the ML models exhibited good 5-year survival calibration. The decision curves for CRC-specific survival at 5 years showed that all the ML models, especially RSF, had higher net benefits than default strategies of treating all or no patients at a range of clinically reasonable risk thresholds. The SHapley Additive exPlanations method revealed that R0 resection, tumor-node-metastasis staging, and the number of positive lymph nodes were important factors for 5-year CRC-specific survival.

CONCLUSIONS

This study showed the potential of applying time-to-event ML predictive algorithms to help predict CRC-specific survival. The RSF, GBM, Cox-Time, and N-MTLR algorithms could provide nonparametric alternatives to the Cox Proportional Hazards model in estimating the survival probability of patients with CRC. The transparent time-to-event ML models help clinicians to more accurately predict the survival rate for these patients and improve patient outcomes by enabling personalized treatment plans that are informed by explainable ML models.

Revolutionizing Oral Cancer Detection: An Approach Using Aquila and Gorilla Algorithms Optimized Transfer Learning-Based CNNs

The early detection of oral cancer is pivotal for improving patient survival rates. However, the high cost of manual initial screenings poses a challenge, especially in resource-limited settings. Deep learning offers an enticing solution by enabling automated and cost-effective screening. This study introduces a groundbreaking empirical framework designed to revolutionize the accurate and automatic classification of oral cancer using microscopic histopathology slide images. This innovative system capitalizes on the power of convolutional neural networks (CNNs), strengthened by the synergy of transfer learning (TL), and further fine-tuned using the novel Aquila Optimizer (AO) and Gorilla Troops Optimizer (GTO), two cutting-edge metaheuristic optimization algorithms. This integration is a novel approach, addressing bias and unpredictability issues commonly encountered in the preprocessing and optimization phases. In the experiments, the capabilities of well-established pre-trained TL models, including VGG19, VGG16, MobileNet, MobileNetV3Small, MobileNetV2, MobileNetV3Large, NASNetMobile, and DenseNet201, all initialized with 'ImageNet' weights, were harnessed. The experimental dataset consisted of the Histopathologic Oral Cancer Detection dataset, which includes a 'normal' class with 2494 images and an 'OSCC' (oral squamous cell carcinoma) class with 2698 images. The results reveal a remarkable performance distinction between the AO and GTO, with the AO consistently outperforming the GTO across all models except for the Xception model. The DenseNet201 model stands out as the most accurate, achieving an astounding average accuracy rate of 99.25% with the AO and 97.27% with the GTO. This innovative framework signifies a significant leap forward in automating oral cancer detection, showcasing the tremendous potential of applying optimized deep learning models in the realm of healthcare diagnostics. The integration of the AO and GTO in our CNN-based system not only pushes the boundaries of classification accuracy but also underscores the transformative impact of metaheuristic optimization techniques in the field of medical image analysis.

Predicting oral cancer survival-Development and validation of an Asia-Pacific nomogram

BACKGROUND

Nomograms are graphical calculating devices that predict response to treatment during cancer management. Oral squamous cell carcinoma (OSCC) is a lethal and deforming disease of rising incidence and global significance. The aim of this study was to develop a nomogram to predict individualized OSCC survival using a population-based dataset obtained from Queensland, Australia and externally validated using a cohort of OSCC patients treated in Hong Kong.

METHODS

Clinico-pathological data for newly diagnosed OSCC patients, including age, sex, tumour site and grading, were accessed retrospectively from the Queensland Cancer Registry (QCR) in Australia and the Clinical Data Analysis and Reporting System (CDARS) in Hong Kong. Multivariate Cox proportional hazard regression was used to construct overall survival (OS) and cancer-specific survival (CSS) prediction models. Nomograms were internally validated using 10-fold cross validation, and externally validated against the Hong Kong dataset.

RESULTS

Data from 9885 OSCC patients in Queensland and 465 patients from Hong Kong were analysed. All clinico-pathological variables significantly influenced survival outcomes. Nomogram calibration curves demonstrated excellent agreement between predicted and actual probability for Queensland patients. External validation in the Hong Kong population demonstrated slightly poorer nomogram performance, but predictive power remained strong.

CONCLUSION

Based upon readily available data documenting patient demographic and clinico-pathological variables, predictive nomograms offer pragmatic aid to clinicians in individualized treatment planning and prognosis assessment in contemporary OSCC management.

Data-centric artificial intelligence in oncology: a systematic review assessing data quality in machine learning models for head and neck cancer

Machine learning models have been increasingly considered to model head and neck cancer outcomes for improved screening, diagnosis, treatment, and prognostication of the disease. As the concept of data-centric artificial intelligence is still incipient in healthcare systems, little is known about the data quality of the models proposed for clinical utility. This is important as it supports the generalizability of the models and data standardization. Therefore, this study overviews the quality of structured and unstructured data used for machine learning model construction in head and neck cancer. Relevant studies reporting on the use of machine learning models based on structured and unstructured custom datasets between January 2016 and June 2022 were sourced from PubMed, EMBASE, Scopus, and Web of Science electronic databases. Prediction model Risk of Bias Assessment (PROBAST) tool was used to assess the quality of individual studies before comprehensive data quality parameters were assessed according to the type of dataset used for model construction. A total of 159 studies were included in the review; 106 utilized structured datasets while 53 utilized unstructured datasets. Data quality assessments were deliberately performed for 14.2% of structured datasets and 11.3% of unstructured datasets before model construction. Class imbalance and data fairness were the most common limitations in data quality for both types of datasets while outlier detection and lack of representative outcome classes were common in structured and unstructured datasets respectively. Furthermore, this review found that class imbalance reduced the discriminatory performance for models based on structured datasets while higher image resolution and good class overlap resulted in better model performance using unstructured datasets during internal validation. Overall, data quality was infrequently assessed before the construction of ML models in head and neck cancer irrespective of the use of structured or unstructured datasets. To improve model generalizability, the assessments discussed in this study should be introduced during model construction to achieve data-centric intelligent systems for head and neck cancer management.

A deep learning-based model predicts survival for patients with laryngeal squamous cell carcinoma: a large population-based study

OBJECTIVES

To assess the performance of DeepSurv, a deep learning-based model in the survival prediction of laryngeal squamous cell carcinoma (LSCC) using the Surveillance, Epidemiology, and End Results (SEER) database.

METHODS

In this large population-based study, we developed and validated a deep learning survival neural network using pathologically diagnosed patients with LSCC from the SEER database between January 2010 and December 2018. Totally 13 variables were included in this network, including patients baseline characteristics, stage, grade, site, tumor extension and treatment details. Based on the total risk score derived from this algorithm, a three-knot restricted cubic spline was plotted to exhibit the difference of survival benefits from two treatment modalities.

RESULTS

Totally 6316 patients with LSCC were included in the study, of which 4237 cases diagnosed between 2010 and 2015 were selected as the development cohort, and the rest (2079 cases diagnosed from 2016 to 2018) were the validation cohort. A state-of-the-art deep learning-based model based on 23 features (i.e., 13 variables) was generated, which showed more superior performance in the prediction of overall survival (OS) than the tumor, node, and metastasis (TNM) staging system (C-index for DeepSurv vs TNM staging = 0.71; 95% CI 0.69-0.74 vs 0.61; 95% CI 0.60-0.63). Interestingly, a significantly nonlinear association between total risk score and treatment effectiveness was observed. When the total risk score ranges 0.1-1.5, surgical treatment brought more survival benefits than nonsurgical one for LSCC patients, especially in 70.5% of patients staged III-IV.

CONCLUSIONS

The deep learning-based model shows more potential benefits in survival estimation for patients with LSCC, which may potentially serve as an auxiliary approach to provide reliable treatment recommendations.

Development and validation of questionnaire-based machine learning models for predicting all-cause mortality in a representative population of China

Background

Considering that the previously developed mortality prediction models have limited applications to the Chinese population, a questionnaire-based prediction model is of great importance for its accuracy and convenience in clinical practice.

Methods

Two national cohort, namely, the China Health and Nutrition Survey (8,355 individual older than 18) and the China Health and Retirement Longitudinal Study (12,711 individuals older than 45) were used for model development and validation. One hundred and fifty-nine variables were compiled to generate predictions. The Cox regression model and six machine learning (ML) models were used to predict all-cause mortality. Finally, a simple questionnaire-based ML prediction model was developed using the best algorithm and validated.

Results

In the internal validation set, all the ML models performed better than the traditional Cox model in predicting 6-year mortality and the random survival forest (RSF) model performed best. The questionnaire-based ML model, which only included 20 variables, achieved a C-index of 0.86 (95%CI: 0.80-0.92). On external validation, the simple questionnaire-based model achieved a C-index of 0.82 (95%CI: 0.77-0.87), 0.77 (95%CI: 0.75-0.79), and 0.79 (95%CI: 0.77-0.81), respectively, in predicting 2-, 9-, and 11-year mortality.

Conclusions

In this prospective population-based study, a model based on the RSF analysis performed best among all models. Furthermore, there was no significant difference between the prediction performance of the questionnaire-based ML model, which only included 20 variables, and that of the model with all variables (including laboratory variables). The simple questionnaire-based ML prediction model, which needs to be further explored, is of great importance for its accuracy and suitability to the Chinese general population.

A systematic review on machine learning and deep learning techniques in cancer survival prediction

Cancer is a disease which is characterised by the unusual and uncontrollable growth of body cells. This usually happens asymptomatically and gets spread to other parts of the body. The major problem in treating cancer is that its progress is not monitored once it is diagnosed. The progress or the prognosis can be done through survival analysis. The survival analysis is the branch of statistics that deals in predicting the time of event of occurrence. In the case of cancer prognosis the event is the survival time of the patient from the onset of the disease or it can be the recurrence of the disease after undergoing a treatment. This study aims to bring out the machine learning and deep learning models involved in providing the prognosis to the cancer patients.

Machine Learning–Based Overall Survival Prediction of Elderly Patients With Multiple Myeloma From Multicentre Real-Life Data

Objective

To use machine learning methods to explore overall survival (OS)-related prognostic factors in elderly multiple myeloma (MM) patients.

Methods

Data were cleaned and imputed using simple imputation methods. Two data resampling methods were implemented to facilitate model building and cross validation. Four algorithms including the cox proportional hazards model (CPH); DeepSurv; DeepHit; and the random survival forest (RSF) were applied to incorporate 30 parameters, such as baseline data, genetic abnormalities and treatment options, to construct a prognostic model for OS prediction in 338 elderly MM patients (>65 years old) from four hospitals in Beijing. The C-index and the integrated Brier score (IBwere used to evaluate model performances.

Results

The 30 variables incorporated in the models comprised MM baseline data, induction treatment data and maintenance therapy data. The variable importance test showed that the OS predictions were largely affected by the maintenance schema variable. Visualizing the survival curves by maintenance schema, we realized that the immunomodulator group had the best survival rate. C-indexes of 0.769, 0.780, 0.785, 0.798 and IBS score of 0.142, 0.112, 0.108, 0.099 were obtained from the CPH model, DeepSurv, DeepHit, and the RSF model respectively. The RSF model yield best scores from the fivefold cross-validation, and the results showed that different data resampling methods did affect our model results.

Conclusion

We established an OS model for elderly MM patients without genomic data based on 30 characteristics and treatment data by machine learning.

Estimating time-to-total knee replacement on radiographs and MRI: a multimodal approach using self-supervised deep learning

Purpose

Accurately predicting the expected duration of time until total knee replacement (time-to-TKR) is crucial for patient management and health care planning. Predicting when surgery may be needed, especially within shorter windows like 3 years, allows clinicians to plan timely interventions and health care systems to allocate resources more effectively. Existing models lack the precision for such time-based predictions. A survival analysis model for predicting time-to-TKR was developed using features from medical images and clinical measurements.

Methods

From the Osteoarthritis Initiative dataset, all knees with clinical variables, MRI scans, radiographs, and quantitative and semiquantitative assessments from images were identified. This resulted in 895 knees that underwent TKR within the 9-year follow-up period, as specified by the Osteoarthritis Initiative study design, and 786 control knees that did not undergo TKR (right-censored, indicating their status beyond the 9-year follow-up is unknown). These knees were used for model training and testing. Additionally, 518 and 164 subjects from the Multi-Center Osteoarthritis Study and internal hospital data were used for external testing, respectively. Deep learning models were utilized to extract features from radiographs and MR scans. Extracted features, clinical variables, and image assessments were used in survival analysis with Lasso Cox feature selection and a random survival forest model to predict time-to-TKR.

Results

The proposed model exhibited strong discrimination power by integrating self-supervised deep learning features with clinical variables (eg, age, body mass index, pain score) and image assessment measurements (eg, Kellgren-Lawrence grade, joint space narrowing, bone marrow lesion size, cartilage morphology) from multiple modalities. The model achieved an area under the curve of 94.5 (95% CI, 94.0-95.1) for predicting the time-to-TKR.

Conclusions

The proposed model demonstrated the potential of self-supervised learning and multimodal data fusion in accurately predicting time-to-TKR that may assist physicians to develop personalize treatment strategies.

Deep Learning Predicts the Malignant-Transformation-Free Survival of Oral Potentially Malignant Disorders

Simple Summary

Mouth cancer is the most common malignancy in the head-and-neck region. Usually, these tumors develop from white lesions in the mouth that appear long before cancer diagnosis. However, platforms that can estimate the time-factored risk of cancer occurring from these diseases and guide treatment and monitoring approaches are elusive. To this end, our study presents time-to-event models that are based on machine learning for prediction of the risk of malignancy from oral white lesions following pathological diagnosis as a function of time. These models displayed very satisfactory discrimination and calibration after multiple tests. To facilitate their preliminary use in clinical practice and further validation, we created a website supporting the use of these models to aid decision making.

Abstract

Machine-intelligence platforms for the prediction of the probability of malignant transformation of oral potentially malignant disorders are required as adjunctive decision-making platforms in contemporary clinical practice. This study utilized time-to-event learning models to predict malignant transformation in oral leukoplakia and oral lichenoid lesions. A total of 1098 patients with oral white lesions from two institutions were included in this study. In all, 26 features available from electronic health records were used to train four learning algorithms—Cox-Time, DeepHit, DeepSurv, random survival forest (RSF)—and one standard statistical method—Cox proportional hazards model. Discriminatory performance, calibration of survival estimates, and model stability were assessed using a concordance index (c-index), integrated Brier score (IBS), and standard deviation of the averaged c-index and IBS following training cross-validation. This study found that DeepSurv (c-index: 0.95, IBS: 0.04) and RSF (c-index: 0.91, IBS: 0.03) were the two outperforming models based on discrimination and calibration following internal validation. However, DeepSurv was more stable than RSF upon cross-validation. External validation confirmed the utility of DeepSurv for discrimination (c-index—0.82 vs. 0.73) and RSF for individual survival estimates (0.18 vs. 0.03). We deployed the DeepSurv model to encourage incipient application in clinical practice. Overall, time-to-event models are successful in predicting the malignant transformation of oral leukoplakia and oral lichenoid lesions.