Machine Learning Decision Tree Models for Differentiation of Posterior Fossa Tumors Using Diffusion Histogram Analysis and Structural MRI Findings

GWO+RuleFit: rule-based explainable machine-learning combined with heuristics to predict mid-treatment FDG PET response to chemoradiation for locally advanced non-small cell lung cancer

Objective.Vital rules learned from fluorodeoxyglucose positron emission tomography (FDG-PET) radiomics of tumor subregional response can provide clinical decision support for precise treatment adaptation. We combined a rule-based machine learning (ML) model (RuleFit) with a heuristic algorithm (gray wolf optimizer, GWO) for mid-chemoradiation FDG-PET response prediction in patients with locally advanced non-small cell lung cancer.Approach.Tumors subregions were identified using K-means clustering. GWO+RuleFit consists of three main parts: (i) a random forest is constructed based on conventional features or radiomic features extracted from tumor regions or subregions in FDG-PET images, from which the initial rules are generated; (ii) GWO is used for iterative rule selection; (iii) the selected rules are fit to a linear model to make predictions about the target variable. Two target variables were considered: a binary response measure (ΔSUVmean ⩾ 20% decline) for classification and a continuous response measure (ΔSUVmean) for regression. GWO+RuleFit was benchmarked against common ML algorithms and RuleFit, with leave-one-out cross-validated performance evaluated by the area under the receiver operating characteristic curve (AUC) in classification and root-mean-square error (RMSE) in regression.Main results.GWO+RuleFit selected 15 rules from the radiomic feature dataset of 23 patients. For treatment response classification, GWO+RuleFit attained numerically better cross-validated performance than RuleFit across tumor regions and sets of features (AUC: 0.58-0.86 vs. 0.52-0.78,p= 0.170-0.925). GWO+Rulefit also had the best or second-best performance numerically compared to all other algorithms for all conditions. For treatment response regression prediction, GWO+RuleFit (RMSE: 0.162-0.192) performed better numerically for low-dimensional models (p= 0.097-0.614) and significantly better for high-dimensional models across all tumor regions except one (RMSE: 0.189-0.219,p< 0.004).Significance. The GWO+RuleFit selected rules were interpretable, highlighting distinct radiomic phenotypes that modulated treatment response. GWO+Rulefit achieved parsimonious models while maintaining utility for treatment response prediction, which can aid clinical decisions for patient risk stratification, treatment selection, and biologically driven adaptation. Clinical trial: NCT02773238.

Role of artificial intelligence in brain tumour imaging

Artificial intelligence (AI) is a rapidly evolving field with many neuro-oncology applications. In this review, we discuss how AI can assist in brain tumour imaging, focusing on machine learning (ML) and deep learning (DL) techniques. We describe how AI can help in lesion detection, differential diagnosis, anatomic segmentation, molecular marker identification, prognostication, and pseudo-progression evaluation. We also cover AI applications in non-glioma brain tumours, such as brain metastasis, posterior fossa, and pituitary tumours. We highlight the challenges and limitations of AI implementation in radiology, such as data quality, standardization, and integration. Based on the findings in the aforementioned areas, we conclude that AI can potentially improve the diagnosis and treatment of brain tumours and provide a path towards personalized medicine and better patient outcomes.

Application of Machine Learning for Classification of Brain Tumors: A Systematic Review and Meta-Analysis

BACKGROUND

Classifying brain tumors accurately is crucial for treatment and prognosis. Machine learning (ML) shows great promise in improving tumor classification accuracy. This study evaluates ML algorithms for differentiating various brain tumor types.

METHODS

A systematic review and meta-analysis were conducted, searching PubMed, Embase, and Web of Science up to March 14, 2023. Studies that only investigated image segmentation accuracy or brain tumor detection instead of classification were excluded. We extracted binary diagnostic accuracy data, constructing contingency tables to derive sensitivity and specificity.

RESULTS

Fifty-one studies were included. The pooled area under the curve for glioblastoma versus lymphoma and low-grade versus high-grade gliomas were 0.99 (95% confidence interval [CI]: 0.98-1.00) and 0.89, respectively. The pooled sensitivity and specificity for benign versus malignant tumors were 0.90 (95% CI: 0.85-0.93) and 0.93 (95% CI: 0.90-0.95), respectively. The pooled sensitivity and specificity for low-grade versus high-grade gliomas were 0.99 (95% CI: 0.97-1.00) and 0.94, (95% CI: 0.79-0.99), respectively. Primary versus metastatic tumor identification yields sensitivity and specificity of 0.89, (95% CI: 0.83-0.93) and 0.87 (95% CI: 0.82-0.91), correspondingly. The differentiation of gliomas from pituitary tumors yielded the highest results among primary brain tumor classifications: sensitivity of 0.99 (95% CI: 0.99-1.00) and specificity of 0.99 (95% CI: 0.98-1.00).

CONCLUSIONS

ML demonstrated excellent performance in classifying brain tumor images, with near-maximum area under the curves, sensitivity, and specificity.

Explainable machine learning predicts survival of retroperitoneal liposarcoma: A study based on the SEER database and external validation in China

OBJECTIVE

We have developed explainable machine learning models to predict the overall survival (OS) of retroperitoneal liposarcoma (RLPS) patients. This approach aims to enhance the explainability and transparency of our modeling results.

METHODS

We collected clinicopathological information of RLPS patients from The Surveillance, Epidemiology, and End Results (SEER) database and allocated them into training and validation sets with a 7:3 ratio. Simultaneously, we obtained an external validation cohort from The First Affiliated Hospital of Naval Medical University (Shanghai, China). We performed LASSO regression and multivariate Cox proportional hazards analysis to identify relevant risk factors, which were then combined to develop six machine learning (ML) models: Cox proportional hazards model (Coxph), random survival forest (RSF), ranger, gradient boosting with component-wise linear models (GBM), decision trees, and boosting trees. The predictive performance of these ML models was evaluated using the concordance index (C-index), the integrated cumulative/dynamic area under the curve (AUC), and the integrated Brier score, as well as the Cox-Snell residual plot. We also used time-dependent variable importance, analysis of partial dependence survival plots, and the generation of aggregated survival SHapley Additive exPlanations (SurvSHAP) plots to provide a global explanation of the optimal model. Additionally, SurvSHAP (t) and survival local interpretable model-agnostic explanations (SurvLIME) plots were used to provide a local explanation of the optimal model.

RESULTS

The final ML models are consisted of six factors: patient's age, gender, marital status, surgical history, as well as tumor's histopathological classification, histological grade, and SEER stage. Our prognostic model exhibits significant discriminative ability, particularly with the ranger model performing optimally. In the training set, validation set, and external validation set, the AUC for 1, 3, and 5 year OS are all above 0.83, and the integrated Brier scores are consistently below 0.15. The explainability analysis of the ranger model also indicates that histological grade, histopathological classification, and age are the most influential factors in predicting OS.

CONCLUSIONS

The ranger ML prognostic model exhibits optimal performance and can be utilized to predict the OS of RLPS patients, offering valuable and crucial references for clinical physicians to make informed decisions in advance.

Applied Artificial Intelligence in Healthcare: A Review of Computer Vision Technology Application in Hospital Settings

Computer vision (CV), a type of artificial intelligence (AI) that uses digital videos or a sequence of images to recognize content, has been used extensively across industries in recent years. However, in the healthcare industry, its applications are limited by factors like privacy, safety, and ethical concerns. Despite this, CV has the potential to improve patient monitoring, and system efficiencies, while reducing workload. In contrast to previous reviews, we focus on the end-user applications of CV. First, we briefly review and categorize CV applications in other industries (job enhancement, surveillance and monitoring, automation, and augmented reality). We then review the developments of CV in the hospital setting, outpatient, and community settings. The recent advances in monitoring delirium, pain and sedation, patient deterioration, mechanical ventilation, mobility, patient safety, surgical applications, quantification of workload in the hospital, and monitoring for patient events outside the hospital are highlighted. To identify opportunities for future applications, we also completed journey mapping at different system levels. Lastly, we discuss the privacy, safety, and ethical considerations associated with CV and outline processes in algorithm development and testing that limit CV expansion in healthcare. This comprehensive review highlights CV applications and ideas for its expanded use in healthcare.

Developing radiology diagnostic tools for pulmonary fibrosis using machine learning methods

BACKGROUND

Accurate and prompt diagnosis of the different patterns for pulmonary fibrosis is essential for patient management. However, accurate diagnosis of the specific pattern is challenging due to overlapping radiographic characteristics.

MATERIALS AND METHODS

We conducted a retrospective chart review utilizing two machine learning methods, classification and regression tree and Bayesian additive regression tree, to select the most important radiographic features for diagnosing the three most common fibrosis patterns and created an online diagnostic app for convenient implementation.

RESULTS

Four hundred patients (median age of 67 with inter quartile range 58-73; 200 males) were included in the study. Peripheral distribution, homogeneity, lower lobe predominance and mosaic attenuation of fibrosis are the four most important features identified. Bayesian additive regression tree demonstrates better performance than classification and regression tree in diagnosis prediction and provides the predicted probability of each diagnosis with uncertainty intervals for each combination of features.

CONCLUSION

The model and app built with Bayesian additive regression tree can be used as an effective tool in assisting radiologists in the diagnostic process of pulmonary fibrosis pattern recognition.

Prediction of lymph node metastasis of lung squamous cell carcinoma by machine learning algorithm classifiers

BACKGROUND

Lymph node metastasis (LNM) is an essential factor affecting the prognosis of patients with lung squamous cell carcinoma (LUSC), as well as a critical consideration for the choice of treatment strategy. Exploring effective methods for predicting LNM in LUSC may benefit clinical decision making.

MATERIALS AND METHODS

We used data collected from the Surveillance, Epidemiology, and End Results (SEER) database to develop machine learning algorithm classifiers, including boosted trees (BTs), based on the primary clinical parameters of patients to predict LNM in LUSC. Training on a large-sample training cohort (n = 8,063) allowed for the construction of several concise classifiers for LNM prediction in LUSC, which were then validated using test and in-house cohorts (n = 2,017 and 57, respectively).

RESULTS

The six classifiers established in this research enabled distinction between patients with and without LNM. Among these classifiers, the BT classifier was the top performer, with accuracy, F1 scores, precision, recall, sensitivity, and specificity values of 0.654, 0.621, 0.654, 0.592, 0.592, and 0.711, respectively. The precision recall (PR) and receiver operating characteristic (ROC) (with area under the curve = 0.714) curves also supported this result, which was validated by the in-house cohort. Notably, the tumor stage was a critical factor in determining LNM in patients with LUSC.

CONCLUSIONS

The use of classifiers, especially the BT classifier, may serve as a useful tool for improving clinical precision and individualized treatment of patients with LUSC.

Deciphering Machine Learning Decisions to Distinguish between Posterior Fossa Tumor Types Using MRI Features: What Do the Data Tell Us?

Machine learning (ML) models have become capable of making critical decisions on our behalf. Nevertheless, due to complexity of these models, interpreting their decisions can be challenging, and humans cannot always control them. This paper provides explanations of decisions made by ML models in diagnosing four types of posterior fossa tumors: medulloblastoma, ependymoma, pilocytic astrocytoma, and brainstem glioma. The proposed methodology involves data analysis using kernel density estimations with Gaussian distributions to examine individual MRI features, conducting an analysis on the relationships between these features, and performing a comprehensive analysis of ML model behavior. This approach offers a simple yet informative and reliable means of identifying and validating distinguishable MRI features for the diagnosis of pediatric brain tumors. By presenting a comprehensive analysis of the responses of the four pediatric tumor types to each other and to ML models in a single source, this study aims to bridge the knowledge gap in the existing literature concerning the relationship between ML and medical outcomes. The results highlight that employing a simplistic approach in the absence of very large datasets leads to significantly more pronounced and explainable outcomes, as expected. Additionally, the study also demonstrates that the pre-analysis results consistently align with the outputs of the ML models and the clinical findings reported in the existing literature.

Machine learning decision tree models for multiclass classification of common malignant brain tumors using perfusion and spectroscopy MRI data

Background

To investigate the contribution of machine learning decision tree models applied to perfusion and spectroscopy MRI for multiclass classification of lymphomas, glioblastomas, and metastases, and then to bring out the underlying key pathophysiological processes involved in the hierarchization of the decision-making algorithms of the models.

Methods

From 2013 to 2020, 180 consecutive patients with histopathologically proved lymphomas (n = 77), glioblastomas (n = 45), and metastases (n = 58) were included in machine learning analysis after undergoing MRI. The perfusion parameters (rCBVmax, PSRmax) and spectroscopic concentration ratios (lac/Cr, Cho/NAA, Cho/Cr, and lip/Cr) were applied to construct Classification and Regression Tree (CART) models for multiclass classification of these brain tumors. A 5-fold random cross validation was performed on the dataset.

Results

The decision tree model thus constructed successfully classified all 3 tumor types with a performance (AUC) of 0.98 for PCNSLs, 0.98 for GBM and 1.00 for METs. The model accuracy was 0.96 with a RSquare of 0.887. Five rules of classifier combinations were extracted with a predicted probability from 0.907 to 0.989 for that end nodes of the decision tree for tumor multiclass classification. In hierarchical order of importance, the root node (Cho/NAA) in the decision tree algorithm was primarily based on the proliferative, infiltrative, and neuronal destructive characteristics of the tumor, the internal node (PSRmax), on tumor tissue capillary permeability characteristics, and the end node (Lac/Cr or Cho/Cr), on tumor energy glycolytic (Warburg effect), or on membrane lipid tumor metabolism.

Conclusion

Our study shows potential implementation of machine learning decision tree model algorithms based on a hierarchical, convenient, and personalized use of perfusion and spectroscopy MRI data for multiclass classification of these brain tumors.

Machine learning for classification of cutaneous sebaceous neoplasms: implementing decision tree model using cytological and architectural features

BACKGROUND

This observational study aims to describe and compare histopathological, architectural, and nuclear characteristics of sebaceous lesions and utilized these characteristics to develop a predictive classification approach using machine learning algorithms.

METHODS

This cross-sectional study was conducted on Iranian patients with sebaceous tumors from two hospitals between March 2015 and March 2019. Pathology slides were reviewed by two pathologists and the architectural and cytological attributes were recorded. Multiple decision tree models were trained using 5-fold cross validation to determine the most important predictor variables and to develop a simple prediction model.

RESULTS

This study assessed the characteristics of 123 sebaceous tumors. Histopathological findings, including pagetoid appearance, neurovascular invasion, atypical mitosis, extensive necrotic area, poor cell differentiation, and non-lobular tumor growth pattern, as well as nuclear features, including highly irregular nuclear contour, and large nuclear size were exclusively observed in carcinomatous tumors. Among non-carcinomatous lesions, some sebaceoma and sebaceous adenoma cases had features like high mitotic activity, which can be misleading and complicate diagnosis. Based on multiple decision tree models, the five most critical variables for lesion categorization were identified as: basaloid cell count, peripheral basaloid cell layers, tumor margin, nuclear size, and chromatin.

CONCLUSIONS

This study implemented a machine learning modeling approach to help optimally categorize sebaceous lesions based on architectural and nuclear features. However, studies of larger sample sizes are needed to ensure the accuracy of our suggested predictive model.

Bone metastasis risk and prognosis assessment models for kidney cancer based on machine learning

Background

Bone metastasis is a common adverse event in kidney cancer, often resulting in poor survival. However, tools for predicting KCBM and assessing survival after KCBM have not performed well.

Methods

The study uses machine learning to build models for assessing kidney cancer bone metastasis risk, prognosis, and performance evaluation. We selected 71,414 kidney cancer patients from SEER database between 2010 and 2016. Additionally, 963 patients with kidney cancer from an independent medical center were chosen to validate the performance. In the next step, eight different machine learning methods were applied to develop KCBM diagnosis and prognosis models while the risk factors were identified from univariate and multivariate logistic regression and the prognosis factors were analyzed through Kaplan-Meier survival curve and Cox proportional hazards regression. The performance of the models was compared with current models, including the logistic regression model and the AJCC TNM staging model, applying receiver operating characteristics, decision curve analysis, and the calculation of accuracy and sensitivity in both internal and independent external cohorts.

Results

Our prognosis model achieved an AUC of 0.8269 (95%CI: 0.8083-0.8425) in the internal validation cohort and 0.9123 (95%CI: 0.8979-0.9261) in the external validation cohort. In addition, we tested the performance of the extreme gradient boosting model through decision curve analysis curve, Precision-Recall curve, and Brier score and two models exhibited excellent performance.

Conclusion

Our developed models can accurately predict the risk and prognosis of KCBM and contribute to helping improve decision-making.

Machine Learning in the Classification of Pediatric Posterior Fossa Tumors: A Systematic Review

Background: Posterior fossa tumors (PFTs) are a morbid group of central nervous system tumors that most often present in childhood. While early diagnosis is critical to drive appropriate treatment, definitive diagnosis is currently only achievable through invasive tissue collection and histopathological analyses. Machine learning has been investigated as an alternative means of diagnosis. In this systematic review and meta-analysis, we evaluated the primary literature to identify all machine learning algorithms developed to classify and diagnose pediatric PFTs using imaging or molecular data. Methods: Of the 433 primary papers identified in PubMed, EMBASE, and Web of Science, 25 ultimately met the inclusion criteria. The included papers were extracted for algorithm architecture, study parameters, performance, strengths, and limitations. Results: The algorithms exhibited variable performance based on sample size, classifier(s) used, and individual tumor types being investigated. Ependymoma, medulloblastoma, and pilocytic astrocytoma were the most studied tumors with algorithm accuracies ranging from 37.5% to 94.5%. A minority of studies compared the developed algorithm to a trained neuroradiologist, with three imaging-based algorithms yielding superior performance. Common algorithm and study limitations included small sample sizes, uneven representation of individual tumor types, inconsistent performance reporting, and a lack of application in the clinical environment. Conclusions: Artificial intelligence has the potential to improve the speed and accuracy of diagnosis in this field if the right algorithm is applied to the right scenario. Work is needed to standardize outcome reporting and facilitate additional trials to allow for clinical uptake.

Radiophysiomics: Brain Tumors Classification by Machine Learning and Physiological MRI Data

Simple Summary

The pretreatment diagnosis of contrast-enhancing brain tumors is still challenging in clinical neuro-oncology due to their very similar appearance on conventional MRI. A precise initial characterization, however, is essential to initiate appropriate treatment management, which can substantially differ between brain tumor entities. To overcome the disadvantage of the low specificity of conventional MRI, several new neuroimaging methods have been developed and validated over the past decades. This increasing amount of diagnostic information makes a timely evaluation without computational support impossible in a clinical setting. Artificial intelligence methods such as machine learning offer new options to support clinicians. In this study, we combined nine common machine learning algorithms with a physiological MRI technique (we named this approach “radiophysiomics”) to investigate the effectiveness of the multiclass classification of contrast-enhancing brain tumors in a clinical setting. We were able to demonstrate that radiophysiomics could be helpful in the routine diagnostics of contrast-enhancing brain tumors, but further automation using deep neural networks is required.

Abstract

The precise initial characterization of contrast-enhancing brain tumors has significant consequences for clinical outcomes. Various novel neuroimaging methods have been developed to increase the specificity of conventional magnetic resonance imaging (cMRI) but also the increased complexity of data analysis. Artificial intelligence offers new options to manage this challenge in clinical settings. Here, we investigated whether multiclass machine learning (ML) algorithms applied to a high-dimensional panel of radiomic features from advanced MRI (advMRI) and physiological MRI (phyMRI; thus, radiophysiomics) could reliably classify contrast-enhancing brain tumors. The recently developed phyMRI technique enables the quantitative assessment of microvascular architecture, neovascularization, oxygen metabolism, and tissue hypoxia. A training cohort of 167 patients suffering from one of the five most common brain tumor entities (glioblastoma, anaplastic glioma, meningioma, primary CNS lymphoma, or brain metastasis), combined with nine common ML algorithms, was used to develop overall 135 classifiers. Multiclass classification performance was investigated using tenfold cross-validation and an independent test cohort. Adaptive boosting and random forest in combination with advMRI and phyMRI data were superior to human reading in accuracy (0.875 vs. 0.850), precision (0.862 vs. 0.798), F-score (0.774 vs. 0.740), AUROC (0.886 vs. 0.813), and classification error (5 vs. 6). The radiologists, however, showed a higher sensitivity (0.767 vs. 0.750) and specificity (0.925 vs. 0.902). We demonstrated that ML-based radiophysiomics could be helpful in the clinical routine diagnosis of contrast-enhancing brain tumors; however, a high expenditure of time and work for data preprocessing requires the inclusion of deep neural networks.

Leveraging MRI characterization of longitudinal tears of the deep digital flexor tendon in horses using machine learning

While MRI is the modality of choice for the diagnosis of longitudinal tears (LTs) of the deep digital flexor tendon (DDFT) of horses, differentiating between various grades of tears based on imaging characteristics is challenging due to overlapping imaging features. In this retrospective, exploratory, diagnostic accuracy study, a machine learning (ML) scheme was applied to link quantitative features and qualitative descriptors to leverage MRI characteristics of different grades of tearing of the DDFT of horses. A qualitative MRI characteristic scheme, combining tendon morphologic features, altered signal intensity, and synovial sheath distention, was used for LT classification with an excellent diagnostic accuracy of the high-grade tears but more limited accuracy for the detection of low-grade tears. A quantitative ML approach was followed to measure the contribution of 30 quantitative phenotypic features for characterizing and classifying tendinous tears. Among the 30 imaging features, boundary curvature represented by the standard deviation and maximum had the most significant discriminatory power (P < 0.05) between normal and abnormal tendons and could be used as an aid for classifying the different grades of LTs of DDFTs. Imaging analysis-based 3D interactive surface plot supports qualitative characterization of different grades of LTs of the DDFT through clearer visualization of the tendon in three dimensions and simple integration of two perspectives features (i.e., margin/distribution and intensity/distribution). A systematic approach combining quantitative features with qualitative analyses using ML was diagnostically beneficial in MRI characterization and in discriminating between different grades of LTs of the DDFT of horses.

Differences in the MRI Signature and ADC Values of Diffuse Midline Gliomas with H3 K27M Mutation Compared to Midline Glioblastomas

We conducted a two-center retrospective survey on standard MRI features including apparent diffusion coefficient mapping (ADC) of diffuse midline gliomas H3 K27M-mutant (DMG) compared to midline glioblastomas H3 K27M-wildtype (midGBM-H3wt). We identified 39 intracranial DMG and 18 midGBM-H3wt tumors. Samples were microscopically re-evaluated for microvascular proliferations and necrosis. Image analysis focused on location, peritumoral edema, degree of contrast enhancement and DWI features. Within DMG, MRI features between tumors with or without histomorphological GBM features were compared. DMG occurred in 15/39 samples from the thalamus (38%), in 23/39 samples from the brainstem (59%) and in 1/39 tumors involving primarily the cerebellum (2%). Edema was present in 3/39 DMG cases (8%) versus 78% in the control (midGBM-H3wt) group (p < 0.001). Contrast enhancement at the tumor rim was detected in 17/39 DMG (44%) versus 67% in control (p = 0.155), and necrosis in 24/39 (62%) versus 89% in control (p = 0.060). Strong contrast enhancement was observed in 15/39 DMG (38%) versus 56% in control (p = 0.262). Apparent diffusion coefficient (ADC) histogram analysis showed significantly higher skewness and kurtosis values in the DMG group compared to the controls (p = 0.0016/p = 0.002). Minimum relative ADC (rADC) values, as well as the 10th and 25th rADC-percentiles, were lower in DMGs with GBM features within the DMG group (p < 0.001/p = 0.012/p = 0.027). In conclusion, DMG cases exhibited markedly less edema than midGBM-H3wt, even if histomorphological malignancy was present. Histologically malignant DMGs and midGBM-H3wt more often displayed strong enhancement, as well as rim enhancement, than DMGs without histomorphological malignancy. DMGs showed higher skewness and kurtosis values on ADC-histogram analysis compared to midGBM-H3wt. Lower minimum rADC values in DMGs indicated malignant histomorphological features, likely representing a more complex tissue microstructure.

Machine Learning Applications for Differentiation of Glioma from Brain Metastasis-A Systematic Review

Simple Summary

We present a systematic review of published reports on machine learning (ML) applications for the differentiation of gliomas from brain metastases by summarizing study characteristics, strengths, and pitfalls. Based on these findings, we present recommendations for future research in this field.

Abstract

Glioma and brain metastasis can be difficult to distinguish on conventional magnetic resonance imaging (MRI) due to the similarity of imaging features in specific clinical circumstances. Multiple studies have investigated the use of machine learning (ML) models for non-invasive differentiation of glioma from brain metastasis. Many of the studies report promising classification results, however, to date, none have been implemented into clinical practice. After a screening of 12,470 studies, we included 29 eligible studies in our systematic review. From each study, we aggregated data on model design, development, and best classifiers, as well as quality of reporting according to the TRIPOD statement. In a subset of eligible studies, we conducted a meta-analysis of the reported AUC. It was found that data predominantly originated from single-center institutions (n = 25/29) and only two studies performed external validation. The median TRIPOD adherence was 0.48, indicating insufficient quality of reporting among surveyed studies. Our findings illustrate that despite promising classification results, reliable model assessment is limited by poor reporting of study design and lack of algorithm validation and generalizability. Therefore, adherence to quality guidelines and validation on outside datasets is critical for the clinical translation of ML for the differentiation of glioma and brain metastasis.

ADC Histogram Analysis of Pediatric Low-Grade Glioma Treated with Selumetinib: A Report from the Pediatric Brain Tumor Consortium

BACKGROUND AND PURPOSE

Selumetinib is a promising MAP (mitogen-activated protein) kinase (MEK) 1/2 inhibitor treatment for pediatric low-grade gliomas. We hypothesized that MR imaging-derived ADC histogram metrics would be associated with survival and response to treatment with selumetinib.

MATERIALS AND METHODS

Children with recurrent, refractory, or progressive pediatric low-grade gliomas who had World Health Organization grade I pilocytic astrocytoma with KIAA1549-BRAF fusion or the BRAF V600E mutation (stratum 1), neurofibromatosis type 1-associated pediatric low-grade gliomas (stratum 3), or sporadic non-neurofibromatosis type 1 optic pathway and hypothalamic glioma (OPHG) (stratum 4) were treated with selumetinib for up to 2 years. Quantitative ADC histogram metrics were analyzed for total and enhancing tumor volumes at baseline and during treatment.

RESULTS

Each stratum comprised 25 patients. Stratum 1 responders showed lower values of SD of baseline ADC_total as well as a larger decrease with time on treatment in ADC_total mean, mode, and median compared with nonresponders. Stratum 3 responders showed a greater longitudinal decrease in ADC_total. In stratum 4, higher baseline ADC_total skewness and kurtosis were associated with shorter progression-free survival. When all 3 strata were combined, responders showed a greater decrease with time in ADC_total mode and median. Compared with sporadic OPHG, neurofibromatosis type 1-associated OPHG had lower values of ADC_total mean, mode, and median as well as ADC_enhancement mean and median and higher values of ADC_total skewness and kurtosis at baseline. The longitudinal decrease in ADC_total median during treatment was significantly greater in sporadic OPHG compared with neurofibromatosis type 1-associated OPHG.

CONCLUSIONS

ADC histogram metrics are associated with progression-free survival and response to treatment with selumetinib in pediatric low-grade gliomas.

Prediction of lung metastases in thyroid cancer using machine learning based on SEER database

PURPOSE

Lung metastasis (LM) is one of the most frequent distant metastases of thyroid cancer (TC). This study aimed to develop a machine learning algorithm model to predict lung metastasis of thyroid cancer for providing relative information in clinical decision-making.

METHODS

Data comprising of demographic and clinicopathological characteristics of patients with thyroid cancer were extracted from the National Institutes of Health (NIH)'s Surveillance, Epidemiology, and End Results (SEER) database between 2010 and 2015, which is employed to develop six machine learning algorithm models support vector machine (SVM), logistic regression (LR), eXtreme gradient boosting (XGBoost), decision tree (DT), random forest (RF), and k-nearest neighbor (KNN). Compared and evaluated models by the following indicators: accuracy, precision, recall rate, F1-score, the area under the ROC curve (AUC) value and Brier score, and interpreted the association between clinicopathological characteristics and target variables based on the best model.

RESULTS

Nine thousand nine hundred and fifty patients were selected, which including 212 patients (2.1%) with lung metastasis, and 9738 patients without lung metastasis (97.9%). Multivariate logistic regression showed that age, T stage, N stage, and histological type were independent factors in TC with LM. Evaluation indicators of the best model- RF were as following: accuracy (0.99), recall rate (0.88), precision (0.61), F1-score (0.72), AUC value (0.99), and the Brier score (0.016).

CONCLUSION

RF learning model performed better and can be applied to forecast lung metastasis of thyroid cancer, and offer valuable and significant reference for clinicians' decision-making in advance.

Beyond Glioma: The Utility of Radiomic Analysis for Non-Glial Intracranial Tumors

Simple Summary

Tumor qualities, such as growth rate, firmness, and intrusion into healthy tissue, can be very important for operation planning and further treatment. Radiomics is a promising new method that allows the determination of some of these qualities on images performed before surgery. In this article, we provide a review of the use of radiomics in various tumors of the central nervous system, such as metastases, lymphoma, meningioma, medulloblastoma, and pituitary tumors.

Abstract

The field of radiomics is rapidly expanding and gaining a valuable role in neuro-oncology. The possibilities related to the use of radiomic analysis, such as distinguishing types of malignancies, predicting tumor grade, determining the presence of particular molecular markers, consistency, therapy response, and prognosis, can considerably influence decision-making in medicine in the near future. Even though the main focus of radiomic analyses has been on glial CNS tumors, studies on other intracranial tumors have shown encouraging results. Therefore, as the main focus of this review, we performed an analysis of publications on PubMed and Web of Science databases, focusing on radiomics in CNS metastases, lymphoma, meningioma, medulloblastoma, and pituitary tumors.

Differentiation of paediatric posterior fossa tumours by the multiregional and multiparametric MRI radiomics approach: a study on the selection of optimal multiple sequences and multiregions

OBJECTIVE

To evaluate the diagnostic performance of a radiomics model based on multiregional and multiparametric MRI to classify paediatric posterior fossa tumours (PPFTs), explore the contribution of different MR sequences and tumour subregions in tumour classification, and examine whether contrast-enhanced T₁ weighted (T1C) images have irreplaceable added value.

METHODS

This retrospective study of 136 PPFTs extracted 11,958 multiregional (enhanced, non-enhanced, and total tumour) features from multiparametric MRI (T₁- and T₂ weighted, T1C, fluid-attenuated inversion recovery, and diffusion-weighted images). These features were subjected to fast correlation-based feature selection and classified by a support vector machine based on different tasks. Diagnostic performances of multiregional and multiparametric MRI features, different sequences, and different tumoral regions were evaluated using multiclass and one-vs-rest strategies.

RESULTS

The established model achieved an overall area under the curve (AUC) of 0.977 in the validation cohort. The performance of PPFTs significantly improved after replacing T1C with apparent diffusion coefficient maps added into the plain scan sequences (AUC from 0.812 to 0.917). When oedema features were added to contrast-enhancing tumour volume, the performance did not significantly improve.

CONCLUSION

The radiomics model built by multiregional and multiparametric MRI features allows for the excellent distinction of different PPFTs and provides valuable references for the rational adoption of MR sequences.

ADVANCES IN KNOWLEDGE

This study emphasized that T1C has limited added value in predicting PPFTs and should be cautiously adopted. Selecting optimal MR sequences may help guide clinicians to better allocate acquisition sequences and reduce medical costs.

Clinical applications of artificial intelligence and radiomics in neuro-oncology imaging

This article is a comprehensive review of the basic background, technique, and clinical applications of artificial intelligence (AI) and radiomics in the field of neuro-oncology. A variety of AI and radiomics utilized conventional and advanced techniques to differentiate brain tumors from non-neoplastic lesions such as inflammatory and demyelinating brain lesions. It is used in the diagnosis of gliomas and discrimination of gliomas from lymphomas and metastasis. Also, semiautomated and automated tumor segmentation has been developed for radiotherapy planning and follow-up. It has a role in the grading, prediction of treatment response, and prognosis of gliomas. Radiogenomics allowed the connection of the imaging phenotype of the tumor to its molecular environment. In addition, AI is applied for the assessment of extra-axial brain tumors and pediatric tumors with high performance in tumor detection, classification, and stratification of patient's prognoses.

Evolving Role and Translation of Radiomics and Radiogenomics in Adult and Pediatric Neuro-Oncology

Exponential technologic advancements in imaging, high-performance computing, and artificial intelligence, in addition to increasing access to vast amounts of diverse data, have revolutionized the role of imaging in medicine. Radiomics is defined as a high-throughput feature-extraction method that unlocks microscale quantitative data hidden within standard-of-care medical imaging. Radiogenomics is defined as the linkage between imaging and genomics information. Multiple radiomics and radiogenomics studies performed on conventional and advanced neuro-oncology image modalities show that they have the potential to differentiate pseudoprogression from true progression, classify tumor subgroups, and predict recurrence, survival, and mutation status with high accuracy. In this article, we outline the technical steps involved in radiomics and radiogenomics analyses with the use of artificial intelligence methods and review current applications in adult and pediatric neuro-oncology.

Ventriculomegaly and postoperative lateral/third ventricular blood as predictors of cerebrospinal fluid diversion following posterior fossa tumor resection

OBJECTIVE

Postoperative hydrocephalus occurs in one-third of children after posterior fossa tumor resection. Although models to predict the need for CSF diversion after resection exist for preoperative variables, it is unknown which postoperative variables predict the need for CSF diversion. In this study, the authors sought to determine the clinical and radiographic predictors for CSF diversion in children following posterior fossa tumor resection.

METHODS

This was a retrospective cohort study involving patients ≤ 18 years of age who underwent resection of a primary posterior fossa tumor between 2000 and 2018. The primary outcome was the need for CSF diversion 6 months after surgery. Candidate predictors for CSF diversion including age, race, sex, frontal occipital horn ratio (FOHR), tumor type, tumor volume and location, transependymal edema, papilledema, presence of postoperative intraventricular blood, and residual tumor were evaluated using a best subset selection method with logistic regression.

RESULTS

Of the 63 included patients, 26 (41.3%) had CSF diversion at 6 months. Patients who required CSF diversion had a higher median FOHR (0.5 vs 0.4) and a higher percentage of postoperative intraventricular blood (30.8% vs 2.7%) compared with those who did not. A 0.1-unit increase in FOHR or intraventricular blood was associated with increased odds of CSF diversion (OR 2.9 [95% CI 1.3-7.8], p = 0.02 and OR 20.2 [95% CI 2.9-423.1], p = 0.01, respectively) with an overfitting-corrected concordance index of 0.68 (95% CI 0.56-0.80).

CONCLUSIONS

The preoperative FOHR and postoperative intraventricular blood were significant predictors of the need for permanent CSF diversion within 6 months after posterior fossa tumor resection in children.

Radiomic Phenotypes Distinguish Atypical Teratoid/Rhabdoid Tumors from Medulloblastoma

BACKGROUND AND PURPOSE

Atypical teratoid/rhabdoid tumors and medulloblastomas have similar imaging and histologic features but distinctly different outcomes. We hypothesized that they could be distinguished by MR imaging-based radiomic phenotypes.

MATERIALS AND METHODS

We retrospectively assembled T2-weighted and gadolinium-enhanced T1-weighted images of 48 posterior fossa atypical teratoid/rhabdoid tumors and 96 match-paired medulloblastomas from 7 institutions. Using a holdout test set, we measured the performance of 6 candidate classifier models using 6 imaging features derived by sparse regression of 900 T2WI and 900 T1WI Imaging Biomarker Standardization Initiative-based radiomics features.

RESULTS

From the originally extracted 1800 total Imaging Biomarker Standardization Initiative-based features, sparse regression consistently reduced the feature set to 1 from T1WI and 5 from T2WI. Among classifier models, logistic regression performed with the highest AUC of 0.86, with sensitivity, specificity, accuracy, and F1 scores of 0.80, 0.82, 0.81, and 0.85, respectively. The top 3 important Imaging Biomarker Standardization Initiative features, by decreasing order of relative contribution, included voxel intensity at the 90th percentile, inverse difference moment normalized, and kurtosis-all from T2WI.

CONCLUSIONS

Six quantitative signatures of image intensity, texture, and morphology distinguish atypical teratoid/rhabdoid tumors from medulloblastomas with high prediction performance across different machine learning strategies. Use of this technique for preoperative diagnosis of atypical teratoid/rhabdoid tumors could significantly inform therapeutic strategies and patient care discussions.

Radiohistogenomics of pediatric low-grade neuroepithelial tumors

PURPOSE

In addition to histology, genetic alteration is now required to classify many central nervous system (CNS) tumors according to the most recent World Health Organization CNS tumor classification scheme. Although that is still not the case for classifying pediatric low-grade neuroepithelial tumors (PLGNTs), genetic and molecular features are increasingly being used for making treatment decisions. This approach has become a standard clinical practice in many specialized pediatric cancer centers and will likely be more widely practiced in the near future. This paradigm shift in the management of PLGNTs necessitates better understanding of how genetic alterations influence histology and imaging characteristics of individual PLGNT phenotypes.

METHODS

The complex association of genetic alterations with histology, clinical, and imaging of each phenotype of the extremely heterogeneous PLGNT family has been addressed in a holistic approach in this up-to-date review article. A new imaging stratification scheme has been proposed based on tumor morphology, location, histology, and genetics. Imaging characteristics of each PLGNT entity are also depicted in light of histology and genetics.

CONCLUSION

This article reviews the association of specific genetic alteration with location, histology, imaging, and prognosis of a specific tumor of the PLGNT family and how that information can be used for better imaging of these tumors.