Development of a Machine Learning Model for Survival Risk Stratification of Patients With Advanced Oral Cancer

Machine learning explainability for survival outcome in head and neck squamous cell carcinoma

BACKGROUND

Diagnosis and treatment of head and neck squamous cell carcinoma (HNSCC) induces psychological variables and treatment-related toxicity in patients. The evaluation of outcomes is warranted for effective treatment planning and improved disease management.

OBJECTIVES

This study aimed to build a prognostic system by combining clinicopathological parameters, treatment-related factors, and sociodemographic factors as integrative inputs to build a machine learning (ML) model to estimate the overall survival (OS) of patients with HNSCC. Furthermore, we explored the complementary prognostic potentials of these input parameters. We provide explainability and interpretability using Local Interpretable Model-agnostic Explanations (LIME) and SHapley Additive exPlanations (SHAP) techniques.

METHODS

A total of 419 patients with HNSCC were recruited from three University Hospitals in Sweden. We compared the performance of TabNet, a state-of-the-art deep learning algorithm for tabular data, with extreme gradient boosting (XGBoost) and voting ensemble to predict OS in patients with HNSCC.

RESULTS

Both TabNet and XGBoost showed comparable performance accuracies, with TabNet and XGBoost showing a performance accuracy of 88.1% each and voting ensemble showing an accuracy of 88.7%. The aggregate feature importance showed that p16 (a tumor suppressor protein that plays a crucial role in cell cycle regulation), cancer stage, hemoglobin, age at diagnosis, T class, N class, smoking pack-years, body mass index (BMI), treatment modality, erythrocyte count, and human papillomavirus (HPV) status were the most important parameters for the predictive ability of the model for OS. Furthermore, we found survival trends in this cohort by individually considering parameters such as p16, cancer stage, hemoglobin, age at diagnosis, HPV status, Tumor Nodal Metastasis staging, and socioeconomic factors (marital status, housing, and level of education). In addition, both the LIME and SHAP techniques showed the contribution of each feature to the prediction made by the model.

CONCLUSIONS

The clinical implementation of an ML model can lead to individualized risk-based therapeutic decision-making. Therefore, validating these models with multi-institutional datasets and testing them in the context of clinical trials is warranted for safe clinical implementation.

MRI in Oral Tongue Squamous Cell Carcinoma: A Radiomic Approach in the Local Recurrence Evaluation

(1) Background: Oral tongue squamous cell carcinoma (OTSCC) is a prevalent malignancy with high loco-regional recurrence. Advanced imaging biomarkers are critical for stratifying patients at a high risk of recurrence. This study aimed to develop MRI-based radiomic models to predict loco-regional recurrence in OTSCC patients undergoing surgery. (2) Methods: We retrospectively selected 92 patients with OTSCC who underwent MRI, followed by surgery and cervical lymphadenectomy. A total of 31 patients suffered from a loco-regional recurrence. Radiomic features were extracted from preoperative post-contrast high-resolution MRI and integrated with clinical and pathological data to develop predictive models, including radiomic-only and combined radiomic-clinical approaches, trained and validated with stratified data splitting. (3) Results: Textural features, such as those derived from the Gray-Level Size-Zone Matrix, Gray-Level Dependence Matrix, and Gray-Level Run-Length Matrix, showed significant associations with recurrence. The radiomic-only model achieved an accuracy of 0.79 (95% confidence interval: 0.69, 0.87) and 0.74 (95% CI: 0.54, 0.89) in the training and validation set, respectively. Combined radiomic and clinical models, incorporating features like the pathological depth of invasion and lymph node status, provided comparable diagnostic performances. (4) Conclusions: MRI-based radiomic models demonstrated the potential for predicting loco-regional recurrence, highlighting their increasingly important role in advancing precision oncology for OTSCC.

Interpretable Machine Learning to Predict the Malignancy Risk of Follicular Thyroid Neoplasms in Extremely Unbalanced Data: Retrospective Cohort Study and Literature Review

Background

Diagnosing and managing follicular thyroid neoplasms (FTNs) remains a significant challenge, as the malignancy risk cannot be determined until after diagnostic surgery.

Objective

We aimed to use interpretable machine learning to predict the malignancy risk of FTNs preoperatively in a real-world setting.

Methods

We conducted a retrospective cohort study at the Peking University Third Hospital in Beijing, China. Patients with postoperative pathological diagnoses of follicular thyroid adenoma (FTA) or follicular thyroid carcinoma (FTC) were included, excluding those without preoperative thyroid ultrasonography. We used 22 predictors involving demographic characteristics, thyroid sonography, and hormones to train 5 machine learning models: logistic regression, least absolute shrinkage and selection operator regression, random forest, extreme gradient boosting, and support vector machine. The optimal model was selected based on discrimination, calibration, interpretability, and parsimony. To address the highly imbalanced data (FTA:FTC ratio>5:1), model discrimination was assessed using both the area under the receiver operating characteristic curve and the area under the precision-recall curve (AUPRC). To interpret the model, we used Shapley Additive Explanations values and partial dependence and individual conditional expectation plots. Additionally, a systematic review was performed to synthesize existing evidence and validate the discrimination ability of the previously developed Thyroid Imaging Reporting and Data System for Follicular Neoplasm scoring criteria to differentiate between benign and malignant FTNs using our data.

Results

The cohort included 1539 patients (mean age 47.98, SD 14.15 years; female: n=1126, 73.16%) with 1672 FTN tumors (FTA: n=1414; FTC: n=258; FTA:FTC ratio=5.5). The random forest model emerged as optimal, identifying mean thyroid-stimulating hormone (TSH) score, mean tumor diameter, mean TSH, TSH instability, and TSH measurement levels as the top 5 predictors in discriminating FTA from FTC, with the area under the receiver operating characteristic curve of 0.79 (95% CI 0.77-0.81) and AUPRC of 0.40 (95% CI 0.37-0.44). Malignancy risk increased nonlinearly with larger tumor diameters and higher TSH instability but decreased nonlinearly with higher mean TSH scores or mean TSH levels. FTCs with small sizes (mean diameter 2.88, SD 1.38 cm) were more likely to be misclassified as FTAs compared to larger ones (mean diameter 3.71, SD 1.36 cm). The systematic review of the 7 included studies revealed that (1) the FTA:FTC ratio varied from 0.6 to 4.0, lower than the natural distribution of 5.0; (2) no studies assessed prediction performance using AUPRC in unbalanced datasets; and (3) external validations of Thyroid Imaging Reporting and Data System for Follicular Neoplasm scoring criteria underperformed relative to the original study.

Conclusions

Tumor size and TSH measurements were important in screening FTN malignancy risk preoperatively, but accurately predicting the risk of small-sized FTNs remains challenging. Future research should address the limitations posed by the extreme imbalance in FTA and FTC distributions in real-world data.

Diagnosis of lymph node metastasis in oral squamous cell carcinoma by an MRI-based deep learning model

BACKGROUND

Cervical lymph node metastasis (LNM) is a well-established poor prognosticator of oral squamous cell carcinoma (OSCC), in which occult metastasis is a subtype that makes prediction challenging. Here, we developed and validated a deep learning (DL) model using magnetic resonance imaging (MRI) for the identification of LNM in OSCC patients.

METHODS

This retrospective diagnostic study developed a three-stage DL model by 45,664 preoperative MRI images from 723 patients in 10 Chinese hospitals between January 2015 and October 2020. It was comprehensively processed from training (8:2), multicenter external validation to reader study. The performance of the DL model was accessed and compared with general and specialized radiologists.

RESULTS

LNM was found in 36.51% of all patients, and the occult metastasis rate was 16.45%. The three-stage DL model together with a random forest classifier achieved the performance in identification of LNM with areas under curve (AUC) of 0.97 (0.93-0.99) in training cohort and AUC of 0.81 (0.74-0.86) in external validation cohorts. The models can reduce the occult metastasis rate up to 89.50% and add more benefit in guiding neck dissection in cN0 patients. DL models tied or exceeded average performance relative to both general and specialized radiologists.

CONCLUSION

Our three-stage DL model based on MRI with three-dimensional sequences was beneficial in detecting LNM and reducing the occult metastasis rate of OSCC patients.

Exploring the role of health-related quality of life measures in predictive modelling for oncology: a systematic review

Health related quality of life (HRQoL) is increasingly assessed in oncology research and routine care, which has led to the inclusion of HRQoL in prediction models. This review aims to describe the current state of oncological prediction models incorporating HRQoL. A systematic literature search for the inclusion of HRQoL in prediction models in oncology was conducted. Selection criteria were a longitudinal study design and inclusion of HRQoL data in prediction models as predictor, outcome, or both. Risk of bias was assessed using the PROBAST tool and quality of reporting was scored with an adapted TRIPOD reporting guideline. From 4747 abstracts, 98 records were included in this review. High risk of bias was found in 71% of the publications. HRQoL was mainly incorporated as predictor (78% (55% predictor only, 23% both predictor and outcome)), with physical functioning and symptom domains selected most frequently as predictor. Few models (23%) predicted HRQoL domains by other or baseline HRQoL domains. HRQoL was used as outcome in 21% of the publications, with a focus on predicting symptoms. There were no difference between AI-based (16%) and classical methods (84%) in model type selection or model performance when using HRQoL data. This review highlights the role of HRQoL as a tool in predicting disease outcomes. Prediction of and with HRQoL is still in its infancy as most of the models are not fully developed. Current models focus mostly on the physical aspects of HRQoL to predict clinical outcomes, and few utilize AI-based methods.

Evaluating generalizability of oncology trial results to real-world patients using machine learning-based trial emulations

Randomized controlled trials (RCTs) evaluating anti-cancer agents often lack generalizability to real-world oncology patients. Although restrictive eligibility criteria contribute to this issue, the role of selection bias related to prognostic risk remains unclear. In this study, we developed TrialTranslator, a framework designed to systematically evaluate the generalizability of RCTs for oncology therapies. Using a nationwide database of electronic health records from Flatiron Health, this framework emulates RCTs across three prognostic phenotypes identified through machine learning models. We applied this approach to 11 landmark RCTs that investigated anti-cancer regimens considered standard of care for the four most prevalent advanced solid malignancies. Our analyses reveal that patients in low-risk and medium-risk phenotypes exhibit survival times and treatment-associated survival benefits similar to those observed in RCTs. In contrast, high-risk phenotypes show significantly lower survival times and treatment-associated survival benefits compared to RCTs. Our results were corroborated by a comprehensive robustness assessment, including examinations of specific patient subgroups, holdout validation and semi-synthetic data simulation. These findings suggest that the prognostic heterogeneity among real-world oncology patients plays a substantial role in the limited generalizability of RCT results. Machine learning frameworks may facilitate individual patient-level decision support and estimation of real-world treatment benefits to guide trial design.

The current landscape of artificial intelligence in oral and maxillofacial surgery- a narrative review

OBJECTIVE

This narrative review aims to explore the current applications and future prospects of AI within the subfields of oral and maxillofacial surgery (OMS), emphasizing its potential benefits and anticipated challenges.

METHODS

A detailed review of the literature was conducted to evaluate the role of AI in oral and maxillofacial surgery. All domains within OMS were reviewed with a focus on diagnostic, therapeutic and prognostic interventions.

RESULTS

AI has been successfully integrated into surgical specialties to enhance clinical outcomes. In OMS, AI demonstrates potential to improve clinical and administrative workflows in both ambulatory and hospital-based settings. Notable applications include more accurate risk prediction, minimally invasive surgical techniques, and optimized postoperative management.

CONCLUSION

OMS stands to benefit enormously from the integration of AI. However, significant roadblocks, such as ethical concerns, data security, and integration challenges, must be addressed to ensure effective adoption. Further research and innovation are needed to fully realize the potential of AI in this specialty.

Holomics and Artificial Intelligence-Driven Precision Oncology for Medullary Thyroid Carcinoma: Addressing Challenges of a Rare and Aggressive Disease

Background/Objective: Medullary thyroid carcinoma (MTC) is a rare yet aggressive form of thyroid cancer comprising a disproportionate share of thyroid cancer-related mortalities, despite its low prevalence. MTC differs from other differentiated thyroid malignancies due to its heterogeneous nature, presenting complexities in both hereditary and sporadic cases. Traditional management guidelines, which are designed primarily for papillary thyroid carcinoma (PTC), fall short in providing the individualized care required for patients with MTC. In recent years, the sheer volume of data generated from clinical evaluations, radiological imaging, pathological assessments, genetic mutations, and immunological profiles has made it humanly impossible for clinicians to simultaneously analyze and integrate these diverse data streams effectively. This data deluge necessitates the adoption of advanced technologies to assist in decision-making processes. Holomics, which is an integrated approach that combines various omics technologies, along with artificial intelligence (AI), emerges as a powerful solution to address these challenges. Methods: This article reviews how AI-driven precision oncology can enhance the diagnostic workup, staging, risk stratification, management, and follow-up care of patients with MTC by processing vast amounts of complex data quickly and accurately. Articles published in English language and indexed in Pubmed were searched. Results: AI algorithms can identify patterns and correlations that may not be apparent to human clinicians, thereby improving the precision of personalized treatment plans. Moreover, the implementation of AI in the management of MTC enables the collation and synthesis of clinical experiences from across the globe, facilitating a more comprehensive understanding of the disease and its treatment outcomes. Conclusions: The integration of holomics and AI in the management of patients with MTC represents a significant advancement in precision oncology. This innovative approach not only addresses the complexities of a rare and aggressive disease but also paves the way for global collaboration and equitable healthcare solutions, ultimately transforming the landscape of treatment and care of patients with MTC. By leveraging AI and holomics, we can strive toward making personalized healthcare accessible to every individual, regardless of their economic status, thereby improving overall survival rates and quality of life for MTC patients worldwide. This global approach aligns with the United Nations Sustainable Development Goal 3, which aims to ensure healthy lives and promote well-being at all ages.

MoCab: A framework for the deployment of machine learning models across health information systems

BACKGROUND AND OBJECTIVE

Machine learning models are vital for enhancing healthcare services. However, integrating them into health information systems (HISs) introduces challenges beyond clinical decision making, such as interoperability and diverse electronic health records (EHR) formats. We proposed Model Cabinet Architecture (MoCab), a framework designed to leverage fast healthcare interoperability resources (FHIR) as the standard for data storage and retrieval when deploying machine learning models across various HISs, addressing the challenges highlighted by platforms such as EPOCH®, ePRISM®, KETOS, and others.

METHODS

The MoCab architecture is designed to streamline predictive modeling in healthcare through a structured framework incorporating several specialized parts. The Data Service Center manages patient data retrieval from FHIR servers. These data are then processed by the Knowledge Model Center, where they are formatted and fed into predictive models. The Model Retraining Center is crucial in continuously updating these models to maintain accuracy in dynamic clinical environments. The framework further incorporates Clinical Decision Support (CDS) Hooks for issuing clinical alerts. It uses Substitutable Medical Apps Reusable Technologies (SMART) on FHIR to develop applications for displaying alerts, prediction results, and patient records.

RESULTS

The MoCab framework was demonstrated using three types of predictive models: a scoring model (qCSI), a machine learning model (NSTI), and a deep learning model (SPC), applied to synthetic data that mimic a major EHR system. The implementations showed how MoCab integrates predictive models with health data for clinical decision support, utilizing CDS Hooks and SMART on FHIR for seamless HIS integration. The demonstration confirmed the practical utility of MoCab in supporting clinical decision making, validated by its application in various healthcare settings.

CONCLUSIONS

We demonstrate MoCab's potential in promoting the interoperability of machine learning models and enhancing its utility across various EHRs. Despite facing challenges like FHIR adoption, MoCab addresses key challenges in adapting machine learning models within healthcare settings, paving the way for further enhancements and broader adoption.

The efficacy and potential mechanisms of pyrotinib in targeting EGFR and HER2 in advanced oral squamous cell carcinoma

BACKGROUND

Human epidermal growth factor receptor 2 (HER2) plays an important role in the progression of multiple solid tumors and induces resistance to epidermal growth factor receptor (EGFR) target treatment. However, the expression status and the clinical significance of HER2 in oral squamous cell carcinoma (OSCC) is still controversial. Pyrotinib (PYR) is a promising novel EGFR/HER2 dual inhibitor, whose efficacy in OSCC has not been determined.

METHODS

57 locally advanced de novo OSCC patients were included in this study to investigate the relationship between the HER2 expression levels and the prognosis by the tissue microarray analysis (TMA). In vitro and in vivo experiments were performed to retrieve the efficacy of PYR in OSCC. The main downstream of HER2 was evaluated by western blotting in OSCC cell lines and xenograft tumors to explore the potential mechanism of PYR.

RESULTS

This study revealed the primary tumor of OSCC had higher HER2 expression levels. Patients with HER2 overexpression had poor overall survival (P < 0.014) and poor disease free survival (P < 0.042). In vitro, PYR suppressed the proliferation, colony formation and migration of OSCC cells. It also promoted apoptosis of OSCC cells and induced cell cycle arrest. Furthermore, PYR was able to inhibit the occurrence and development of OSCC effectively in vivo. Western blotting revealed that PYR suppressed OSCC by inhibiting the phosphorylation of HER2, AKT and ERK.

CONCLUSIONS

This study exhibited the anti-OSCC effects of PYR in vitro and in vivo, and demonstrated PYR inhibited OSCC cells by inducing apoptosis via the HER2/ AKT and ERK pathway. The result of this study also indicated locally advanced OSCC patients might benefit from HER2 assay and EGFR/HER2 dual inhibit treatment.

Development of a Folding Paper System To Enable the Analysis of Gene Profile of Short- and Long-Distance Cancer Cell Migration

In cancer metastasis, where mortality rates remain high despite advancements in medical treatments, understanding the molecular pathways and cellular dynamics underlying tumor spread is critical for devising more effective therapeutic strategies. Here, a folding paper system was proposed and developed to mimic native tumor microenvironment. This system, composed of 7 stacked layers of paper enclosed in a holder, allows for the culture of cancer cells under conditions mimicking those found in solid tumors, including limited oxygen and nutrients. Because of the migratory capabilities of cancer cells, the cells in the center layer could migrated to outer layers of the paper stack, enabling the differentiation of cells based on their migratory potential. Subsequent gene expression analysis, conducted through RT-PCR and RNA sequencing, revealed significant correlations between cancer cell migration distance and the expression of genes associated with hypoxia, metabolism, ATP production, and cellular process. Moreover, our study identified cells with aggressive phenotypic traits from the outer layers of the paper stack, highlighting the potential of this system for enabling the study of aggressive cancer cell characteristics. Validation of the folding paper system against clinical carcinoma tissue demonstrated its ability to faithfully mimic the native tumor microenvironment. Overall, our findings underscore the utility of the folding paper system as a valuable tool for investigating and identifying critical molecular pathways involved in cancer metastasis.

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.

Multi-omics based artificial intelligence for cancer research

With significant advancements of next generation sequencing technologies, large amounts of multi-omics data, including genomics, epigenomics, transcriptomics, proteomics, and metabolomics, have been accumulated, offering an unprecedented opportunity to explore the heterogeneity and complexity of cancer across various molecular levels and scales. One of the promising aspects of multi-omics lies in its capacity to offer a holistic view of the biological networks and pathways underpinning cancer, facilitating a deeper understanding of its development, progression, and response to treatment. However, the exponential growth of data generated by multi-omics studies present significant analytical challenges. Processing, analyzing, integrating, and interpreting these multi-omics datasets to extract meaningful insights is an ambitious task that stands at the forefront of current cancer research. The application of artificial intelligence (AI) has emerged as a powerful solution to these challenges, demonstrating exceptional capabilities in deciphering complex patterns and extracting valuable information from large-scale, intricate omics datasets. This review delves into the synergy of AI and multi-omics, highlighting its revolutionary impact on oncology. We dissect how this confluence is reshaping the landscape of cancer research and clinical practice, particularly in the realms of early detection, diagnosis, prognosis, treatment and pathology. Additionally, we elaborate the latest AI methods for multi-omics integration to provide a comprehensive insight of the complex biological mechanisms and inherent heterogeneity of cancer. Finally, we discuss the current challenges of data harmonization, algorithm interpretability, and ethical considerations. Addressing these challenges necessitates a multidisciplinary collaboration, paving the promising way for more precise, personalized, and effective treatments for cancer patients.

The efficacy of the National Surgical Quality Improvement Program surgical risk calculator in head and neck surgery: A meta-analysis

BACKGROUND

The National Surgical Quality Improvement Program surgical risk calculator (SRC) estimates the risk for postoperative complications. This meta-analysis assesses the efficacy of the SRC in the field of head and neck surgery.

METHODS

A systematic review identified studies comparing the SRC's predictions to observed outcomes following head and neck surgeries. Predictive accuracy was assessed using receiver operating characteristic curves (AUCs) and Brier scoring.

RESULTS

Nine studies totaling 1774 patients were included. The SRC underpredicted the risk of all outcomes (including any complication [observed (ob) = 35.9%, predicted (pr) = 21.8%] and serious complication [ob = 28.7%, pr = 17.0%]) except mortality (ob = 0.37%, pr = 1.55%). The observed length of stay was more than twice the predicted length (p < 0.02). Discrimination was acceptable for postoperative pneumonia (AUC = 0.778) and urinary tract infection (AUC = 0.782) only. Predictive accuracy was low for all outcomes (Brier scores ≥0.01) and comparable for patients with and without free-flap reconstructions.

CONCLUSION

The SRC is an ineffective instrument for predicting outcomes in head and neck surgery.

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.

A machine learning-based prediction model of pelvic lymph node metastasis in women with early-stage cervical cancer

OBJECTIVE

To develop a novel machine learning-based preoperative prediction model for pelvic lymph node metastasis (PLNM) in early-stage cervical cancer by combining the clinical findings and preoperative computerized tomography (CT) of the whole abdomen and pelvis.

METHODS

Patients diagnosed with International Federation of Gynecology and Obstetrics stage IA2-IIA1 squamous cell carcinoma, adenocarcinoma, and adenosquamous carcinoma of the cervix who had primary radical surgery with bilateral pelvic lymphadenectomy from January 1, 2003 to December 31, 2020, were included. Seven supervised machine learning algorithms, including logistic regression, random forest, support vector machine, adaptive boosting, gradient boosting, extreme gradient boosting, and category boosting, were used to evaluate the risk of PLNM.

RESULTS

PLNM was found in 199 (23.9%) of 832 patients included. Younger age, larger tumor size, higher stage, no prior conization, tumor appearance, adenosquamous histology, and vaginal metastasis as well as the CT findings of larger tumor size, parametrial metastasis, pelvic lymph node enlargement, and vaginal metastasis, were significantly associated with PLNM. The models' predictive performance, including accuracy (89.1%-90.6%), area under the receiver operating characteristics curve (86.9%-91.0%), sensitivity (77.4%-82.4%), specificity (92.1%-94.3%), positive predictive value (77.0%-81.7%), and negative predictive value (93.0%-94.4%), appeared satisfactory and comparable among all the algorithms. After optimizing the model's decision threshold to enhance the sensitivity to at least 95%, the 'highly sensitive' model was obtained with a 2.5%-4.4% false-negative rate of PLNM prediction.

CONCLUSION

We developed prediction models for PLNM in early-stage cervical cancer with promising prediction performance in our setting. Further external validation in other populations is needed with potential clinical applications.

Integrating Artificial Intelligence for Advancing Multiple-Cancer Early Detection via Serum Biomarkers: A Narrative Review

The concept and policies of multicancer early detection (MCED) have gained significant attention from governments worldwide in recent years. In the era of burgeoning artificial intelligence (AI) technology, the integration of MCED with AI has become a prevailing trend, giving rise to a plethora of MCED AI products. However, due to the heterogeneity of both the detection targets and the AI technologies, the overall diversity of MCED AI products remains considerable. The types of detection targets encompass protein biomarkers, cell-free DNA, or combinations of these biomarkers. In the development of AI models, different model training approaches are employed, including datasets of case-control studies or real-world cancer screening datasets. Various validation techniques, such as cross-validation, location-wise validation, and time-wise validation, are used. All of the factors show significant impacts on the predictive efficacy of MCED AIs. After the completion of AI model development, deploying the MCED AIs in clinical practice presents numerous challenges, including presenting the predictive reports, identifying the potential locations and types of tumors, and addressing cancer-related information, such as clinical follow-up and treatment. This study reviews several mature MCED AI products currently available in the market, detecting their composing factors from serum biomarker detection, MCED AI training/validation, and the clinical application. This review illuminates the challenges encountered by existing MCED AI products across these stages, offering insights into the continued development and obstacles within the field of MCED AI.

Prediction of postoperative recurrence of oral cancer by artificial intelligence model: Multilayer perceptron

BACKGROUND

Postoperative recurrence of oral cancer is an important factor affecting the prognosis of patients. Artificial intelligence is used to establish a machine learning model to predict the risk of postoperative recurrence of oral cancer.

METHODS

The information of 387 patients with postoperative oral cancer were collected to establish the multilayer perceptron (MLP) model. The comprehensive variable model was compared with the characteristic variable model, and the MLP model was compared with other models to evaluate the sensitivity of different models in the prediction of postoperative recurrence of oral cancer.

RESULTS

The overall performance of the MLP model under comprehensive variable input was the best.

CONCLUSION

The MLP model has good sensitivity to predict postoperative recurrence of oral cancer, and the predictive model with variable input training is better than that with characteristic variable input.

The Use of Artificial Intelligence Algorithms in the Prognosis and Detection of Lymph Node Involvement in Head and Neck Cancer and Possible Impact in the Development of Personalized Therapeutic Strategy: A Systematic Review

Given the increasingly important role that the use of artificial intelligence algorithms is taking on in the medical field today (especially in oncology), the purpose of this systematic review is to analyze the main reports on such algorithms applied for the prognostic evaluation of patients with head and neck malignancies. The objective of this paper is to examine the currently available literature in the field of artificial intelligence applied to head and neck oncology, particularly in the prognostic evaluation of the patient with this kind of tumor, by means of a systematic review. The paper exposes an overview of the applications of artificial intelligence in deriving prognostic information related to the prediction of survival and recurrence and how these data may have a potential impact on the choice of therapeutic strategy, making it increasingly personalized. This systematic review was written following the PRISMA 2020 guidelines.

Application of machine learning model in predicting the likelihood of blood transfusion after hip fracture surgery

OBJECTIVE

Anemia is one of the common adverse reactions after hip fracture surgery. The traditional method to solve anemia is allogeneic transfusion. However, the transfusion may lead to some complications such as septicemia and fever. So far, few studies have reported roles of machine learning in predicting whether blood transfusion is needed or not after hip fracture surgery. Therefore, the purpose of this study is to develop machine learning models to predict the likelihood of postoperative blood transfusion in patients undergoing hip fracture surgery.

METHODS

This study enrolled 1355 patients who underwent hip fracture surgery at the Affiliated Hospital of Qingdao University from January 2016 to December 2021. Among all patients, 210 cases received postoperative blood transfusion. All patients were randomly divided into a training group and a testing group at a ratio of 7:3. In the training group, univariate and multivariate logistic regression analyses were used to determine independent risk factors for the postoperative transfusion. Then, based on these independent risk factors, tenfold cross-validation method was utilized to develop five machine learning models, including logistic, multilayer perceptron (MLP), extreme gradient boosting (XGBoost), random forest (RF), and support vector machine (SVM). The receiver operating characteristic (ROC) curve, area under ROC curve (AUC), and Matthews correlation coefficient (MCC) were generated to evaluate the performance of the models. Calibration plot and decision curve analysis (DCA) were used to test the performance, stability, and clinical applicability of the models. The models were validated using the testing group; and the ROC curve, MCC, calibration plot, and DCA curves were also generated to validate the performance, stability, and clinical applicability of the models. To further verify the robustness of the model, we randomly grabbed 70% of the samples in the testing set, performed 1000 iterations, and calculated the AUC and confidence interval of the five models. Finally, we used SHapley Additive exPlanations (SHAP) to explain these models.

RESULTS

Multivariate logistic regression analysis showed that there were 8 independent risk factors, including age, blood transfusion history, albumin (ALB), globulin (GLO), total bilirubin (TBIL), indirect bilirubin (IBIL), hemoglobin (HB), and blood loss > 200 ml. We finally selected five independent risk factors including HB, GLO, age, IBIL, and blood loss > 200 ml. Based on these five independent risk factors, we generated six characteristic variables, namely HB, HB × HB, HB × blood loss, GLO × HB, age, age × IBIL, and established five machine learning models using a tenfold cross-validation method. In the training group, the AUC values of logistic, RF, MLP, SVM, and XGB were 0.9320, 0.8911, 0.9327, 0.9225, and 0.8825, respectively, and the average AUC was 0.9122 ± 0.0212. The MCC values were 0.65, 0.77, 0.65, 0.66, and 0.68, respectively, and the calibration plot and DCA performed well. In the testing group the AUC values of logistic, RF, MLP, SVM, and XGB were 0.8483, 0.7978, 0.8576, 0.8598, and 0.8216, respectively. The average AUC was 0.8370 ± 0.0238, and the MCC values were 0.41, 0.35, 0.40, 0.41, and 0.41, respectively. The calibration plot and DCA in the testing group also showed good performance. The AUC values and confidence intervals of the 1000-iteration model were: logistic (AUC, min confidence interval [CI]-max confidence interval [CI] 0.848, 0.804-0.903), RF (AUC, minCI-maxCI 0.797, 0.734-0.857), MLP (AUC, minCI-maxCI 0.858, 0.812-0.902), SVM (AUC, minCI-maxCI 0.859, 0.819-0.910), and XGB (AUC, minCI-maxCI 0.821, 0.764-0.894). The model performed well. Finally, according to SHAP, among all five models, HB played the most important role in model prediction and interpretation.

CONCLUSION

The five models we developed all performed well in predicting the likelihood of blood transfusion after hip fracture surgery. Therefore, we believed that the prediction model based on machine learning had great application prospects in clinical practice, which could help clinicians better predict the risk of blood transfusion after hip fracture surgery.

Application of Machine Learning in Predicting Hepatic Metastasis or Primary Site in Gastroenteropancreatic Neuroendocrine Tumors

Gastroenteropancreatic neuroendocrine tumors (GEP-NETs) account for 80% of gastroenteropancreatic neuroendocrine neoplasms (GEP-NENs). GEP-NETs are well-differentiated tumors, highly heterogeneous in biology and origin, and are often diagnosed at the metastatic stage. Diagnosis is commonly through clinical symptoms, histopathology, and PET-CT imaging, while molecular markers for metastasis and the primary site are unknown. Here, we report the identification of multi-gene signatures for hepatic metastasis and primary sites through analyses on RNA-SEQ datasets of pancreatic and small intestinal NETs tissue samples. Relevant gene features, identified from the normalized RNA-SEQ data using the mRMRe algorithm, were used to develop seven Machine Learning models (LDA, RF, CART, k-NN, SVM, XGBOOST, GBM). Two multi-gene random forest (RF) models classified primary and metastatic samples with 100% accuracy in training and test cohorts and >90% accuracy in an independent validation cohort. Similarly, three multi-gene RF models identified the pancreas or small intestine as the primary site with 100% accuracy in training and test cohorts, and >95% accuracy in an independent cohort. Multi-label models for concurrent prediction of hepatic metastasis and primary site returned >98.42% and >87.42% accuracies on training and test cohorts, respectively. A robust molecular signature to predict liver metastasis or the primary site for GEP-NETs is reported for the first time and could complement the clinical management of GEP-NETs.

lab: an R package for generating analysis-ready data from laboratory records

Background

Electronic health records (EHRs) play a crucial role in healthcare decision-making by giving physicians insights into disease progression and suitable treatment options. Within EHRs, laboratory test results are frequently utilized for predicting disease progression. However, processing laboratory test results often poses challenges due to variations in units and formats. In addition, leveraging the temporal information in EHRs can improve outcomes, prognoses, and diagnosis predication. Nevertheless, the irregular frequency of the data in these records necessitates data preprocessing, which can add complexity to time-series analyses.

Methods

To address these challenges, we developed an open-source R package that facilitates the extraction of temporal information from laboratory records. The proposed lab package generates analysis-ready time series data by segmenting the data into time-series windows and imputing missing values. Moreover, users can map local laboratory codes to the Logical Observation Identifier Names and Codes (LOINC), an international standard. This mapping allows users to incorporate additional information, such as reference ranges and related diseases. Moreover, the reference ranges provided by LOINC enable us to categorize results into normal or abnormal. Finally, the analysis-ready time series data can be further summarized using descriptive statistics and utilized to develop models using machine learning technologies.

Results

Using the lab package, we analyzed data from MIMIC-III, focusing on newborns with patent ductus arteriosus (PDA). We extracted time-series laboratory records and compared the differences in test results between patients with and without 30-day in-hospital mortality. We then identified significant variations in several laboratory test results 7 days after PDA diagnosis. Leveraging the time series-analysis-ready data, we trained a prediction model with the long short-term memory algorithm, achieving an area under the receiver operating characteristic curve of 0.83 for predicting 30-day in-hospital mortality in model training. These findings demonstrate the lab package's effectiveness in analyzing disease progression.

Conclusions

The proposed lab package simplifies and expedites the workflow involved in laboratory records extraction. This tool is particularly valuable in assisting clinical data analysts in overcoming the obstacles associated with heterogeneous and sparse laboratory records.

Prediction of 5-year overall survival of tongue cancer based machine learning

OBJECTIVE

We aimed to develop a 5-year overall survival prediction model for patients with oral tongue squamous cell carcinoma based on machine learning methods.

SUBJECTS AND METHODS

The data were obtained from electronic medical records of 224 OTSCC patients at the PLA General Hospital. A five-year overall survival prediction model was constructed using logistic regression, Support Vector Machines, Decision Tree, Random Forest, Extreme Gradient Boosting, and Light Gradient Boosting Machine. Model performance was evaluated according to the area under the curve (AUC) of the receiver operating characteristic curve. The output of the optimal model was explained using the Python package (SHapley Additive exPlanations, SHAP).

RESULTS

After passing through the grid search and secondary modeling, the Light Gradient Boosting Machine was the best prediction model (AUC = 0.860). As explained by SHapley Additive exPlanations, N-stage, age, systemic inflammation response index, positive lymph nodes, plasma fibrinogen, lymphocyte-to-monocyte ratio, neutrophil percentage, and T-stage could perform a 5-year overall survival prediction for OTSCC. The 5-year survival rate was 42%.

CONCLUSION

The Light Gradient Boosting Machine prediction model predicted 5-year overall survival in OTSCC patients, and this predictive tool has potential prognostic implications for patients with OTSCC.

Risk Stratification for Herpes Simplex Virus Pneumonia Using Elastic Net Penalized Cox Proportional Hazard Algorithm with Enhanced Explainability

Herpes simplex virus (HSV) pneumonia is a serious and often fatal respiratory tract infection that occurs in immunocompromised individuals. The early detection of accurate risk stratification is essential in identifying patients who are at high risk of mortality and may benefit from more aggressive treatment. In this study, we developed and validated a risk stratification model for HSV bronchopneumonia using an elastic net penalized Cox proportional hazard algorithm. We analyzed data from a cohort of 104 critically ill patients with HSV bronchopneumonia identified in Chang Gung Memorial Hospital, Linkou, Taiwan: one of the largest tertiary medical centers in the world. A total of 109 predictors, both clinical and laboratory, were identified in this process to develop a risk stratification model that could accurately predict mortality in patients with HSV bronchopneumonia. This model was able to differentiate the risk of death and predict mortality in patients with HSV bronchopneumonia compared to the APACHE II score in the early stage of ICU admissions. Both hazard ratio coefficient and selection frequency were used as the metrics to enhance the explainability of the informative predictors. Our findings suggest that the elastic net penalized Cox proportional hazard algorithm is a promising tool for risk stratification in patients with HSV bronchopneumonia and could be useful in identifying those at high risk of mortality.

The use of artificial intelligence models to predict survival in patients with laryngeal squamous cell carcinoma

Most recent survival prediction has been based on TNM staging, which does not provide individualized information. However, clinical factors including performance status, age, sex, and smoking might influence survival. Therefore, we used artificial intelligence (AI) to analyze various clinical factors to precisely predict the survival of patients with larynx squamous cell carcinoma (LSCC). We included patients with LSCC (N = 1026) who received definitive treatment from 2002 to 2020. Age, sex, smoking, alcohol consumption, Eastern Cooperative Oncology Group (ECOG) performance status, location of tumor, TNM stage, and treatment methods were analyzed using deep neural network (DNN) with multi-classification and regression, random survival forest (RSF), and Cox proportional hazards (COX-PH) model for prediction of overall survival. Each model was confirmed with five-fold cross validation, and performance was evaluated using linear slope, y-intercept, and C-index. The DNN with multi-classification model demonstrated the highest prediction power (1.000 ± 0.047, 0.126 ± 0.762, and 0.859 ± 0.018 for slope, y-intercept, and C-index, respectively), and the prediction survival curve showed the strongest agreement with the validation survival curve, followed by DNN with regression (0.731 ± 0.048, 9.659 ± 0.964, and 0.893 ± 0.017, respectively). The DNN model produced with only T/N staging showed the poorest survival prediction. When predicting the survival of LSCC patients, various clinical factors should be considered. In the present study, DNN with multi-class was shown to be an appropriate method for survival prediction. AI analysis may predict survival more accurately and improve oncologic outcomes.

The added values of 18F-FDG PET/CT in differentiating cancer recurrence and osteoradionecrosis of mandible in patients with treated oral squamous cell carcinoma

BACKGROUND

Osteoradionecrosis (ORN) of the jaw requires a differential diagnosis to exclude cancer recurrence. Here, we sought to develop a scoring system comprising ¹⁸F-FDG PET/CT parameters for distinguishing between the two conditions in patients with oral squamous cell carcinoma (OSCC).

METHODS

The study consisted of 103 OSCC patients with suspected ORN of the jaw. All participants underwent ¹⁸F-FDG PET/CT imaging within 6 months of diagnostic histopathology. Following extraction of PET parameters, we identified clinical and imaging predictors of mandibular recurrence-free survival (MRFS) using receiver operating characteristic curve analysis and multivariate Cox regression models.

RESULTS

The results of histopathology revealed mandibular cancer recurrence in 24 patients (23.3%). Multivariate Cox regression analyses identified an age at diagnosis ≤ 52 years (P = 0.013), a location of the SUVmax voxel with soft tissue predominance (P = 0.019), and mandibular total lesion glycolysis (TLG) > 62.68 g (P < 0.001) as independent risk factors for MRFS. A scoring system was devised with scores from 0 (no risk factor) to 3 (presence of all three risk factors). High-risk patients with a score of 2-3 compared with score of 0-1 had a significantly higher likelihood of mandibular cancer recurrence (hazard ratio: 32.50, 95% confidence interval: 8.51-124.18, P < 0.001). The scoring system had a sensitivity of 87.50%, a specificity of 82.28%, and an accuracy of 83.50% for identifying mandibular cancer recurrence.

CONCLUSIONS

The scoring system of our study is clinically useful for identifying mandibular cancer recurrence in patients with suspected ORN of the jaw.

Machine learning in metastatic cancer research: Potentials, possibilities, and prospects

Cancer has received extensive recognition for its high mortality rate, with metastatic cancer being the top cause of cancer-related deaths. Metastatic cancer involves the spread of the primary tumor to other body organs. As much as the early detection of cancer is essential, the timely detection of metastasis, the identification of biomarkers, and treatment choice are valuable for improving the quality of life for metastatic cancer patients. This study reviews the existing studies on classical machine learning (ML) and deep learning (DL) in metastatic cancer research. Since the majority of metastatic cancer research data are collected in the formats of PET/CT and MRI image data, deep learning techniques are heavily involved. However, its black-box nature and expensive computational cost are notable concerns. Furthermore, existing models could be overestimated for their generality due to the non-diverse population in clinical trial datasets. Therefore, research gaps are itemized; follow-up studies should be carried out on metastatic cancer using machine learning and deep learning tools with data in a symmetric manner.

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.

Artificial intelligence to predict outcomes of head and neck radiotherapy

Head and neck radiotherapy induces important toxicity, and its efficacy and tolerance vary widely across patients. Advancements in radiotherapy delivery techniques, along with the increased quality and frequency of image guidance, offer a unique opportunity to individualize radiotherapy based on imaging biomarkers, with the aim of improving radiation efficacy while reducing its toxicity. Various artificial intelligence models integrating clinical data and radiomics have shown encouraging results for toxicity and cancer control outcomes prediction in head and neck cancer radiotherapy. Clinical implementation of these models could lead to individualized risk-based therapeutic decision making, but the reliability of the current studies is limited. Understanding, validating and expanding these models to larger multi-institutional data sets and testing them in the context of clinical trials is needed to ensure safe clinical implementation. This review summarizes the current state of the art of machine learning models for prediction of head and neck cancer radiotherapy outcomes.

Empirical comparison of routinely collected electronic health record data for head and neck cancer-specific survival in machine-learnt prognostic models

BACKGROUND

Knowledge of the prognostic factors and performance of machine learning predictive models for 2-year cancer-specific survival (CSS) is limited in the head and neck cancer (HNC) population.

METHODS

Data from our facilities' oncology information system (OIS) collected for routine practice (OIS dataset, n = 430 patients) and research purposes (research dataset, n = 529 patients) were extracted on adults diagnosed between 2000 and 2017 with squamous cell carcinoma of the head and neck.

RESULTS

Machine learning demonstrated excellent performance (area under the curve, AUC) in the whole cohort (AUC = 0.97, research dataset), larynx cohort (AUC = 0.98, both datasets), and oropharynx cohort (AUC = 0.99, both datasets). Tumor site and T classification were identified as predictors of 2-year CSS in both datasets. Hypothyroidism and fitness for operation were further identified in the research dataset.

CONCLUSIONS

Datasets extracted from an OIS for routine clinical practice and research purposes demonstrated high utility for informing 2-year head and neck CSS.

Machine Learning and Nomogram Prognostic Modeling for 2-Year Head and Neck Cancer-Specific Survival Using Electronic Health Record Data: A Multisite Study

PURPOSE

There is limited knowledge of the prediction of 2-year cancer-specific survival (CSS) in the head and neck cancer (HNC) population. The aim of this study is to develop and validate machine learning models and a nomogram for the prediction of 2-year CSS in patients with HNC using real-world data collected by major teaching and tertiary referral hospitals in New South Wales (NSW), Australia.

MATERIALS AND METHODS

Data collected in oncology information systems at multiple NSW Cancer Centres were extracted for 2,953 eligible adults diagnosed between 2000 and 2017 with squamous cell carcinoma of the head and neck. Death data were sourced from the National Death Index using record linkage. Machine learning and Cox regression/nomogram models were developed and internally validated in Python and R, respectively.

RESULTS

Machine learning models demonstrated highest performance (C-index) in the larynx and nasopharynx cohorts (0.82), followed by the oropharynx (0.79) and the hypopharynx and oral cavity cohorts (0.73). In the whole HNC population, C-indexes of 0.79 and 0.70 and Brier scores of 0.10 and 0.27 were reported for the machine learning and nomogram model, respectively. Cox regression analysis identified age, T and N classification, and time-corrected biologic equivalent dose in two gray fractions as independent prognostic factors for 2-year CSS. N classification was the most important feature used for prediction in the machine learning model followed by age.

CONCLUSION

Machine learning and nomogram analysis predicted 2-year CSS with high performance using routinely collected and complete clinical information extracted from oncology information systems. These models function as visual decision-making tools to guide radiotherapy treatment decisions and provide insight into the prediction of survival outcomes in patients with HNC.

Applying interpretable machine learning algorithms to predict risk factors for permanent stoma in patients after TME

Objective

The purpose of this study was to develop a machine learning model to identify preoperative and intraoperative high-risk factors and to predict the occurrence of permanent stoma in patients after total mesorectal excision (TME).

Methods

A total of 1,163 patients with rectal cancer were included in the study, including 142 patients with permanent stoma. We collected 24 characteristic variables, including patient demographic characteristics, basic medical history, preoperative examination characteristics, type of surgery, and intraoperative information. Four machine learning algorithms including extreme gradient boosting (XGBoost), random forest (RF), support vector machine (SVM) and k-nearest neighbor algorithm (KNN) were applied to construct the model and evaluate the model using k-fold cross validation method, ROC curve, calibration curve, decision curve analysis (DCA) and external validation.

Results

The XGBoost algorithm showed the best performance among the four prediction models. The ROC curve results showed that XGBoost had a high predictive accuracy with an AUC value of 0.987 in the training set and 0.963 in the validation set. The k-fold cross-validation method was used for internal validation, and the XGBoost model was stable. The calibration curves showed high predictive power of the XGBoost model. DCA curves showed higher benefit rates for patients who received interventional treatment under the XGBoost model. The AUC value for the external validation set was 0.89, indicating that the XGBoost prediction model has good extrapolation.

Conclusion

The prediction model for permanent stoma in patients with rectal cancer derived from the XGBoost machine learning algorithm in this study has high prediction accuracy and clinical utility.

Identification of High-Risk Patients for Postoperative Myocardial Injury After CME Using Machine Learning: A 10-Year Multicenter Retrospective Study

Purpose

The occurrence of myocardial injury, a grave complication post complete mesocolic excision (CME), profoundly impacts the immediate and long-term prognosis of patients. The aim of this inquiry was to conceive a machine learning model that can recognize preoperative, intraoperative and postoperative high-risk factors and predict the onset of myocardial injury following CME.

Patients and Methods

This study included 1198 colon cancer patients, 133 of whom experienced myocardial injury after surgery. Thirty-six distinct variables were gathered, encompassing patient demographics, medical history, preoperative examination characteristics, surgery type, and intraoperative details. Four machine learning algorithms, namely, extreme gradient boosting (XGBoost), random forest (RF), multilayer perceptron (MLP), and k-nearest neighbor algorithm (KNN), were employed to fabricate the model, and k-fold cross-validation, ROC curve, calibration curve, decision curve analysis (DCA), and external validation were employed to evaluate it.

Results

Out of the four predictive models employed, the XGBoost algorithm demonstrated the best performance. The ROC curve findings indicated that the XGBoost model exhibited remarkable predictive accuracy, with an area under the curve (AUC) value of 0.997 in the training set and 0.956 in the validation set. For internal validation, the k-fold cross-validation method was utilized, and the XGBoost model was shown to be steady. Furthermore, the calibration curves demonstrated the XGBoost model's high predictive capability. The DCA curve revealed higher benefit rates for patients who underwent interventional treatment under the XGBoost model. The AUC value for the external validation set was 0.74, which indicated that the XGBoost prediction model possessed good extrapolative capacity.

Conclusion

The myocardial injury prediction model for patients undergoing CME that was developed using the XGBoost machine learning algorithm in this study demonstrates both high predictive accuracy and clinical utility.

Predicting progression-free survival after systemic therapy in advanced head and neck cancer: Bayesian regression and model development

Background

Advanced head and neck squamous cell carcinoma (HNSCC) is associated with a poor prognosis, and biomarkers that predict response to treatment are highly desirable. The primary aim was to predict progression-free survival (PFS) with a multivariate risk prediction model.

Methods

Experimental covariates were derived from blood samples of 56 HNSCC patients which were prospectively obtained within a Phase 2 clinical trial (NCT02633800) at baseline and after the first treatment cycle of combined platinum-based chemotherapy with cetuximab treatment. Clinical and experimental covariates were selected by Bayesian multivariate regression to form risk scores to predict PFS.

Results

A 'baseline' and a 'combined' risk prediction model were generated, each of which featuring clinical and experimental covariates. The baseline risk signature has three covariates and was strongly driven by baseline percentage of CD33+CD14+HLADRhigh monocytes. The combined signature has six covariates, also featuring baseline CD33+CD14+HLADRhigh monocytes but is strongly driven by on-treatment relative change of CD8+ central memory T cells percentages. The combined model has a higher predictive power than the baseline model and was successfully validated to predict therapeutic response in an independent cohort of nine patients from an additional Phase 2 trial (NCT03494322) assessing the addition of avelumab to cetuximab treatment in HNSCC. We identified tissue counterparts for the immune cells driving the models, using imaging mass cytometry, that specifically colocalized at the tissue level and correlated with outcome.

Conclusions

This immune-based combined multimodality signature, obtained through longitudinal peripheral blood monitoring and validated in an independent cohort, presents a novel means of predicting response early on during the treatment course.

Funding

Daiichi Sankyo Inc, Cancer Research UK, EU IMI2 IMMUCAN, UK Medical Research Council, European Research Council (335326), Merck Serono. Cancer Research Institute, National Institute for Health Research, Guy's and St Thomas' NHS Foundation Trust and The Institute of Cancer Research.

Clinical trial number

NCT02633800.

Machine Learning Heuristics on Gingivobuccal Cancer Gene Datasets Reveals Key Candidate Attributes for Prognosis

Delayed cancer detection is one of the common causes of poor prognosis in the case of many cancers, including cancers of the oral cavity. Despite the improvement and development of new and efficient gene therapy treatments, very little has been carried out to algorithmically assess the impedance of these carcinomas. In this work, from attributes or NCBI's oral cancer datasets, viz. (i) name, (ii) gene(s), (iii) protein change, (iv) condition(s), clinical significance (last reviewed). We sought to train the number of instances emerging from them. Further, we attempt to annotate viable attributes in oral cancer gene datasets for the identification of gingivobuccal cancer (GBC). We further apply supervised and unsupervised machine learning methods to the gene datasets, revealing key candidate attributes for GBC prognosis. Our work highlights the importance of automated identification of key genes responsible for GBC that could perhaps be easily replicated in other forms of oral cancer detection.

Development and validation of machine learning-based risk prediction models of oral squamous cell carcinoma using salivary autoantibody biomarkers

INTRODUCTION

The incidence of oral cavity squamous cell carcinoma (OSCC) continues to rise. OSCC is associated with a low average survival rate, and most patients have a poor disease prognosis because of delayed diagnosis. We used machine learning techniques to predict high-risk cases of OSCC by using salivary autoantibody levels and demographic and behavioral data.

METHODS

We collected the salivary samples of patients recruited from a teaching hospital between September 2008 and December 2012. Ten salivary autoantibodies, sex, age, smoking, alcohol consumption, and betel nut chewing were used to build prediction models for identifying patients with a high risk of OSCC. The machine learning algorithms applied in the study were logistic regression, random forest, support vector machine with the radial basis function kernel, eXtreme Gradient Boosting (XGBoost), and a stacking model. We evaluated the performance of the models by using the area under the receiver operating characteristic curve (AUC), with simulations conducted 100 times.

RESULTS

A total of 337 participants were enrolled in this study. The best predictive model was constructed using a stacking algorithm with original forms of age and logarithmic levels of autoantibodies (AUC = 0.795 ± 0.055). Adding autoantibody levels as a data source significantly improved the prediction capability (from 0.698 ± 0.06 to 0.795 ± 0.055, p < 0.001).

CONCLUSIONS

We successfully established a prediction model for high-risk cases of OSCC. This model can be applied clinically through an online calculator to provide additional personalized information for OSCC diagnosis, thereby reducing the disease morbidity and mortality rates.

Cell marathon: long-distance cell migration and metastasis-associated gene analysis using a folding paper system

A high mortality rate was found in cancer patients with distant metastasis. Development of targeted drugs for effectively inhibiting cancer metastasis is the key for increasing therapeutic success. In the current study, a folding paper system was developed to mimic a tumor-vascular interface for the study of long-distance cell migration. Correlation between the cell migration distance and metastasis-associated gene was successfully analyzed by disassembling the stacked paper construct. The result revealed that the migration distance and number of migrated cells were highly correlated to cell characteristics. Moreover, immunocytochemistry was directly conducted on the paper layer to study the signaling pathway. Kelch-like and protein tyrosine phosphatase families were examined and the PTPN13 gene was shown to regulate long-distance cell migration. By analyzing the phosphorylated mTOR, the PTPN13 gene was further confirmed to be a tumor suppressor gene that inhibits long-distance cell migration. The folding paper system provides an alternative approach for long-distance cell migration. Metastasis-associated gene expression can be analyzed to potentially develop targeted drugs for cancer metastasis inhibition.

Mixture survival trees for cancer risk classification

In oncology studies, it is important to understand and characterize disease heterogeneity among patients so that patients can be classified into different risk groups and one can identify high-risk patients at the right time. This information can then be used to identify a more homogeneous patient population for developing precision medicine. In this paper, we propose a mixture survival tree approach for direct risk classification. We assume that the patients can be classified into a pre-specified number of risk groups, where each group has distinct survival profile. Our proposed tree-based methods are devised to estimate latent group membership using an EM algorithm. The observed data log-likelihood function is used as the splitting criterion in recursive partitioning. The finite sample performance is evaluated by extensive simulation studies and the proposed method is illustrated by a case study in breast cancer.

Sepsis Related Mortality Associated with an Inflammatory Burst in Patients Admitting to the Department of Internal Medicine with Apparently Normal C-Reactive Protein Concentration

Background: Patients who are admitted to the Department of Internal Medicine with apparently normal C-reactive protein (CRP) concentration impose a special challenge due the assumption that they might not harbor a severe and potentially lethal medical condition. Methods: A retrospective cohort of all patients who were admitted to the Department of Internal Medicine with a CRP concentration of ≤31.9 mg/L and had a second CRP test obtained within the next 24 h. Seven day mortality data were analyzed. Results: Overall, 3504 patients were analyzed with a mean first and second CRP of 8.8 (8.5) and 14.6 (21.6) mg/L, respectively. The seven day mortality increased from 1.8% in the first quartile of the first CRP to 7.5% in the fourth quartile of the first CRP (p < 0.0001) and from 0.6% in the first quartile of the second CRP to 9.5% in the fourth quartile of the second CRP test (p < 0.0001), suggesting a clear relation between the admission CRP and in hospital seven day mortality. Conclusions: An association exists between the quartiles of CRP and 7-day mortality as well as sepsis related cause of death. Furthermore, the CRP values 24 h after hospital admission improved the discrimination.

Cardiovascular Autonomic Function Changes and Predictors During a 2-Year Physical Activity Program in Rheumatoid Arthritis: A PARA 2010 Substudy

Background: Chronic inflammation leads to autonomic dysfunction, which may contribute to the increased risk of cardiovascular diseases (CVD) in patients with rheumatoid arthritis (RA). Exercise is known to restore autonomic nervous system (ANS) activity and particularly its parasympathetic component. A practical clinical tool to assess autonomic function, and in particular parasympathetic tone, is heart rate recovery (HRR). The aim of this substudy from the prospective PARA 2010 study was to determine changes in HRR post-maximal exercise electrocardiogram (ECG) after a 2-year physical activity program and to determine the main predictive factors associated with effects on HRR in RA. Methods: Twenty-five participants performed physiotherapist-guided aerobic and muscle-strengthening exercises for 1 year and were instructed to continue the unsupervised physical activity program autonomously in the next year. All participants were examined at baseline and at years 1 and 2 with a maximal exercise ECG on a cycle ergometer. HRR was measured at 1, 2, 3, 4, and 5 min following peak heart rate during exercise. Machine-learning algorithms with the elastic net linear regression models were performed to predict changes in HRR1 and HRR2 at 1 year and 2 years of the PARA program. Results: Mean age was 60 years, range of 41-73 years (88% women). Both HRR1 and HRR2 increased significantly from baseline to year 1 with guided physical activity and decreased significantly from year 1 to year 2 with unsupervised physical activity. Blood pressure response to exercise, low BMI, and muscular strength were the best predictors of HRR1/HRR2 increase during the first year and HRR1/HRR2 decrease during the second year of the PARA program. Conclusion: ANS activity in RA assessed by HRR was improved by guided physical activity, and machine learning allowed to identify predictors of the HRR response at the different time points. HRR could be a relevant marker of the effectiveness of physical activity recommended in patients with RA at high risk of CVD. Very inactive and/or high CVD risk RA patients may get substantial benefits from a physical activity program.

Applications of artificial intelligence in dentistry: A comprehensive review

OBJECTIVE

To perform a comprehensive review of the use of artificial intelligence (AI) and machine learning (ML) in dentistry, providing the community with a broad insight on the different advances that these technologies and tools have produced, paying special attention to the area of esthetic dentistry and color research.

MATERIALS AND METHODS

The comprehensive review was conducted in MEDLINE/PubMed, Web of Science, and Scopus databases, for papers published in English language in the last 20 years.

RESULTS

Out of 3871 eligible papers, 120 were included for final appraisal. Study methodologies included deep learning (DL; n = 76), fuzzy logic (FL; n = 12), and other ML techniques (n = 32), which were mainly applied to disease identification, image segmentation, image correction, and biomimetic color analysis and modeling.

CONCLUSIONS

The insight provided by the present work has reported outstanding results in the design of high-performance decision support systems for the aforementioned areas. The future of digital dentistry goes through the design of integrated approaches providing personalized treatments to patients. In addition, esthetic dentistry can benefit from those advances by developing models allowing a complete characterization of tooth color, enhancing the accuracy of dental restorations.

CLINICAL SIGNIFICANCE

The use of AI and ML has an increasing impact on the dental profession and is complementing the development of digital technologies and tools, with a wide application in treatment planning and esthetic dentistry procedures.

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

BACKGROUND

Applying machine learning to predicting oral cavity cancer prognosis is important in selecting candidates for aggressive treatment following diagnosis. However, models proposed so far have only considered cancer survival as discrete rather than dynamic outcomes.

OBJECTIVES

To compare the model performance of different machine learning-based algorithms that incorporate time-to-event data. These algorithms included DeepSurv, DeepHit, neural net-extended time-dependent cox model (Cox-Time), and random survival forest (RSF).

MATERIALS AND METHODS

Retrospective cohort of 313 oral cavity cancer patients were obtained from electronic health records. Models were trained on patient data following preprocessing. Predictors were based on demographic, clinicopathologic, and treatment information of the cases. Outcomes were the disease-specific and overall survival. Multivariable analyses were conducted to select significant prognostic features associated with tumor prognosis. Two models were generated per algorithm based on all-prognostic features and significant-prognostic features following statistical analysis. Concordance index (c-index) and integrated Brier scores were used as performance evaluators and model stability was assessed using intraclass correlation coefficients (ICC) calculated from these measures obtained from the cross-validation folds.

RESULTS

While all models were satisfactory, better discriminatory performance and calibration was observed for disease-specific than overall survival (mean c-index: 0.85 vs 0.74; mean integrated Brier score: 0.12 vs 0.17). DeepSurv performed best in terms of discrimination for both outcomes (c-indices: 0.76 -0.89) while RSF produced better calibrated survival estimates (integrated Brier score: 0.06 -0.09). Model stability of the algorithms varied with the outcomes as Cox-Time had the best intraclass correlation coefficient (mean ICC: 1.00) for disease-specific survival while DeepSurv was most stable for overall survival prediction (mean ICC: 0.99).

CONCLUSIONS

Machine learning algorithms based on time-to-event outcomes are successful in predicting oral cavity cancer prognosis with DeepSurv and RSF producing the best discriminative performance and calibration.

Machine Learning for Head and Neck Cancer: A Safe Bet?-A Clinically Oriented Systematic Review for the Radiation Oncologist

BACKGROUND AND PURPOSE

Machine learning (ML) is emerging as a feasible approach to optimize patients' care path in Radiation Oncology. Applications include autosegmentation, treatment planning optimization, and prediction of oncological and toxicity outcomes. The purpose of this clinically oriented systematic review is to illustrate the potential and limitations of the most commonly used ML models in solving everyday clinical issues in head and neck cancer (HNC) radiotherapy (RT).

MATERIALS AND METHODS

Electronic databases were screened up to May 2021. Studies dealing with ML and radiomics were considered eligible. The quality of the included studies was rated by an adapted version of the qualitative checklist originally developed by Luo et al. All statistical analyses were performed using R version 3.6.1.

RESULTS

Forty-eight studies (21 on autosegmentation, four on treatment planning, 12 on oncological outcome prediction, 10 on toxicity prediction, and one on determinants of postoperative RT) were included in the analysis. The most common imaging modality was computed tomography (CT) (40%) followed by magnetic resonance (MR) (10%). Quantitative image features were considered in nine studies (19%). No significant differences were identified in global and methodological scores when works were stratified per their task (i.e., autosegmentation).

DISCUSSION AND CONCLUSION

The range of possible applications of ML in the field of HN Radiation Oncology is wide, albeit this area of research is relatively young. Overall, if not safe yet, ML is most probably a bet worth making.

Accessing Artificial Intelligence for Clinical Decision-Making

Advancements in computing and data from the near universal acceptance and implementation of electronic health records has been formative for the growth of personalized, automated, and immediate patient care models that were not previously possible. Artificial intelligence (AI) and its subfields of machine learning, reinforcement learning, and deep learning are well-suited to deal with such data. The authors in this paper review current applications of AI in clinical medicine and discuss the most likely future contributions that AI will provide to the healthcare industry. For instance, in response to the need to risk stratify patients, appropriately cultivated and curated data can assist decision-makers in stratifying preoperative patients into risk categories, as well as categorizing the severity of ailments and health for non-operative patients admitted to hospitals. Previous overt, traditional vital signs and laboratory values that are used to signal alarms for an acutely decompensating patient may be replaced by continuously monitoring and updating AI tools that can pick up early imperceptible patterns predicting subtle health deterioration. Furthermore, AI may help overcome challenges with multiple outcome optimization limitations or sequential decision-making protocols that limit individualized patient care. Despite these tremendously helpful advancements, the data sets that AI models train on and develop have the potential for misapplication and thereby create concerns for application bias. Subsequently, the mechanisms governing this disruptive innovation must be understood by clinical decision-makers to prevent unnecessary harm. This need will force physicians to change their educational infrastructure to facilitate understanding AI platforms, modeling, and limitations to best acclimate practice in the age of AI. By performing a thorough narrative review, this paper examines these specific AI applications, limitations, and requisites while reviewing a few examples of major data sets that are being cultivated and curated in the US.

Prediction models applying machine learning to oral cavity cancer outcomes: A systematic review

OBJECTIVES

Machine learning platforms are now being introduced into modern oncological practice for classification and prediction of patient outcomes. To determine the current status of the application of these learning models as adjunctive decision-making tools in oral cavity cancer management, this systematic review aims to summarize the accuracy of machine-learning based models for disease outcomes.

METHODS

Electronic databases including PubMed, Scopus, EMBASE, Cochrane Library, LILACS, SciELO, PsychINFO, and Web of Science were searched up until December 21, 2020. Pertinent articles detailing the development and accuracy of machine learning prediction models for oral cavity cancer outcomes were selected in a two-stage process. Quality assessment was conducted using the Quality in Prognosis Studies (QUIPS) tool and results of base studies were qualitatively synthesized by all authors. Outcomes of interest were malignant transformation of precancer lesions, cervical lymph node metastasis, as well as treatment response, and prognosis of oral cavity cancer.

RESULTS

Twenty-seven articles out of 950 citations identified from electronic and manual searching were included in this study. Five studies had low bias concerns on the QUIPS tool. Prediction of malignant transformation, cervical lymph node metastasis, treatment response, and prognosis were reported in three, six, eight, and eleven articles respectively. Accuracy of these learning models on the internal or external validation sets ranged from 0.85 to 0.97 for malignant transformation prediction, 0.78-0.91 for cervical lymph node metastasis prediction, 0.64-1.00 for treatment response prediction, and 0.71-0.99 for prognosis prediction. In general, most trained algorithms predicting these outcomes performed better than alternate methods of prediction. We also found that models including molecular markers in training data had better accuracy estimates for malignant transformation, treatment response, and prognosis prediction.

CONCLUSION

Machine learning algorithms have a satisfactory to excellent accuracy for predicting three of four oral cavity cancer outcomes i.e., malignant transformation, nodal metastasis, and prognosis. However, considering the training approach of many available classifiers, these models may not be streamlined enough for clinical application currently.

iPMI: Machine Learning-Aided Identification of Parametrial Invasion in Women with Early-Stage Cervical Cancer

Radical hysterectomy is a recommended treatment for early-stage cervical cancer. However, the procedure is associated with significant morbidities resulting from the removal of the parametrium. Parametrial cancer invasion (PMI) is found in a minority of patients but the efficient system used to predict it is lacking. In this study, we develop a novel machine learning (ML)-based predictive model based on a random forest model (called iPMI) for the practical identification of PMI in women. Data of 1112 stage IA-IIA cervical cancer patients who underwent primary surgery were collected and considered as the training dataset, while data from an independent cohort of 116 consecutive patients were used as the independent test dataset. Based on these datasets, iPMI-Econ was then developed by using basic clinicopathological data available prior to surgery, while iPMI-Power was also introduced by adding pelvic node metastasis and uterine corpus invasion to the iPMI-Econ. Both 10-fold cross-validations and independent test results showed that iPMI-Power outperformed other well-known ML classifiers (e.g., logistic regression, decision tree, k-nearest neighbor, multi-layer perceptron, naive Bayes, support vector machine, and extreme gradient boosting). Upon comparison, it was found that iPMI-Power was effective and had a superior performance to other well-known ML classifiers in predicting PMI. It is anticipated that the proposed iPMI may serve as a cost-effective and rapid approach to guide important clinical decision-making.

Current Insights into Oral Cancer Diagnostics

Oral cancer is one of the most common head and neck malignancies and has an overall 5-year survival rate that remains below 50%. Oral cancer is generally preceded by oral potentially malignant disorders (OPMDs) but determining the risk of OPMD progressing to cancer remains a difficult task. Several diagnostic technologies have been developed to facilitate the detection of OPMD and oral cancer, and some of these have been translated into regulatory-approved in vitro diagnostic systems or medical devices. Furthermore, the rapid development of novel biomarkers, electronic systems, and artificial intelligence may help to develop a new era where OPMD and oral cancer are detected at an early stage. To date, a visual oral examination remains the routine first-line method of identifying oral lesions; however, this method has certain limitations and as a result, patients are either diagnosed when their cancer reaches a severe stage or a high-risk patient with OPMD is misdiagnosed and left untreated. The purpose of this article is to review the currently available diagnostic methods for oral cancer as well as possible future applications of novel promising technologies to oral cancer diagnosis. This will potentially increase diagnostic options and improve our ability to effectively diagnose and treat oral cancerous-related lesions.

Laboratory Demand Management Strategies-An Overview

Inappropriate laboratory test selection in the form of overutilization as well as underutilization frequently occurs despite available guidelines. There is broad approval among laboratory specialists as well as clinicians that demand management strategies are useful tools to avoid this issue. Most of these tools are based on automated algorithms or other types of machine learning. This review summarizes the available demand management strategies that may be adopted to local settings. We believe that artificial intelligence may help to further improve these available tools.

Computing the Hazard Ratios Associated With Explanatory Variables Using Machine Learning Models of Survival Data

PURPOSE

The application of Cox proportional hazards (CoxPH) models to survival data and the derivation of hazard ratio (HR) are well established. Although nonlinear, tree-based machine learning (ML) models have been developed and applied to the survival analysis, no methodology exists for computing HRs associated with explanatory variables from such models. We describe a novel way to compute HRs from tree-based ML models using the SHapley Additive exPlanation values, which is a locally accurate and consistent methodology to quantify explanatory variables' contribution to predictions.

METHODS

We used three sets of publicly available survival data consisting of patients with colon, breast, or pan cancer and compared the performance of CoxPH with the state-of-the-art ML model, XGBoost. To compute the HR for explanatory variables from the XGBoost model, the SHapley Additive exPlanation values were exponentiated and the ratio of the means over the two subgroups was calculated. The CI was computed via bootstrapping the training data and generating the ML model 1,000 times. Across the three data sets, we systematically compared HRs for all explanatory variables. Open-source libraries in Python and R were used in the analyses.

RESULTS

For the colon and breast cancer data sets, the performance of CoxPH and XGBoost was comparable, and we showed good consistency in the computed HRs. In the pan-cancer data set, we showed agreement in most variables but also an opposite finding in two of the explanatory variables between the CoxPH and XGBoost result. Subsequent Kaplan-Meier plots supported the finding of the XGBoost model.

CONCLUSION

Enabling the derivation of HR from ML models can help to improve the identification of risk factors from complex survival data sets and to enhance the prediction of clinical trial outcomes.