A generic support vector machine model for preoperative glioma survival associations

Magnetic resonance imaging-based radiomics analysis for prediction of treatment response to neoadjuvant chemoradiotherapy and clinical outcome in patients with locally advanced rectal cancer: A large multicentric and validated study

Our study investigated whether magnetic resonance imaging (MRI)-based radiomics features could predict good response (GR) to neoadjuvant chemoradiotherapy (nCRT) and clinical outcome in patients with locally advanced rectal cancer (LARC). Radiomics features were extracted from the T2 weighted (T2W) and Apparent diffusion coefficient (ADC) images of 1070 LARC patients retrospectively and prospectively recruited from three hospitals. To create radiomic models for GR prediction, three classifications were utilized. The radiomic model with the best performance was integrated with important clinical MRI features to create the combined model. Finally, two clinical MRI features and ten radiomic features were chosen for GR prediction. The combined model, constructed with the tumor size, MR-detected extramural venous invasion, and radiomic signature generated by Support Vector Machine (SVM), showed promising discrimination of GR, with area under the curves of 0.799 (95% CI, 0.760-0.838), 0.797 (95% CI, 0.733-0.860), 0.754 (95% CI, 0.678-0.829), and 0.727 (95% CI, 0.641-0.813) in the training and three validation datasets, respectively. Decision curve analysis verified the clinical usefulness. Furthermore, according to Kaplan-Meier curves, patients with a high likelihood of GR as determined by the combined model had better disease-free survival than those with a low probability. This radiomics model was developed based on large-sample size, multicenter datasets, and prospective validation with high radiomics quality score, and also had clinical utility.

Machine Learning in Neurosurgery: Toward Complex Inputs, Actionable Predictions, and Generalizable Translations

Machine learning can predict neurosurgical diagnosis and outcomes, power imaging analysis, and perform robotic navigation and tumor labeling. State-of-the-art models can reconstruct and generate images, predict surgical events from video, and assist in intraoperative decision-making. In this review, we will detail the neurosurgical applications of machine learning, ranging from simple to advanced models, and their potential to transform patient care. As machine learning techniques, outputs, and methods become increasingly complex, their performance is often more impactful yet increasingly difficult to evaluate. We aim to introduce these advancements to the neurosurgical audience while suggesting major potential roadblocks to their safe and effective translation. Unlike the previous generation of machine learning in neurosurgery, the safe translation of recent advancements will be contingent on neurosurgeons' involvement in model development and validation.

A Combined Model Integrating Radiomics and Deep Learning Based on Contrast-Enhanced CT for Preoperative Staging of Laryngeal Carcinoma

RATIONALE AND OBJECTIVES

Accurate staging of laryngeal carcinoma can inform appropriate treatment decision-making. We developed a radiomics model, a deep learning (DL) model, and a combined model (incorporating radiomics features and DL features) based on the venous-phase CT images and explored the performance of these models in stratifying patients with laryngeal carcinoma into stage I-II and stage III-IV, and also compared these models with radiologists.

MATERIALS AND METHODS

Three hundreds and nineteen patients with pathologically confirmed laryngeal carcinoma were randomly divided into a training set (n = 223) and a test set (n = 96). In the training set, the radiomics features with inter- and intraclass correlation coefficients (ICCs)> 0.75 were screened by Spearman correlation analysis and recursive feature elimination (RFE); then support vector machine (SVM) classifier was applied to develop the radiomics model. The DL model was built using ResNet 18 by the cropped 2D regions of interest (ROIs) in the maximum tumor ROI slices and the last fully connected layer of this network served as the DL feature extractor. Finally, a combined model was developed by pooling the radiomics features and extracted DL features to predict the staging.

RESULTS

The area under the curves (AUCs) for radiomics model, DL model, and combined model in the test set were 0.704 (95% confidence interval [CI]: 0.588-0.820), 0.724 (95% CI: 0.613-0.835), and 0.849 (95% CI: 0.755-0.943), respectively. The combined model outperformed the radiomics model and the DL model in discriminating stage I-II from stage III-IV (p = 0.031 and p = 0.020, respectively). Only the combined model performed significantly better than radiologists (p < 0.050 for both).

CONCLUSION

The combined model can help tailor the therapeutic strategy for laryngeal carcinoma patients by enabling more accurate preoperative staging.

Neurosurgical Management of Cerebrospinal Tumors in the Era of Artificial Intelligence : A Scoping Review

Central nervous system tumors are identified as tumors of the brain and spinal cord. The associated morbidity and mortality of cerebrospinal tumors are disproportionately high compared to other malignancies. While minimally invasive techniques have initiated a revolution in neurosurgery, artificial intelligence (AI) is expediting it. Our study aims to analyze AI's role in the neurosurgical management of cerebrospinal tumors. We conducted a scoping review using the Arksey and O'Malley framework. Upon screening, data extraction and analysis were focused on exploring all potential implications of AI, classification of these implications in the management of cerebrospinal tumors. AI has enhanced the precision of diagnosis of these tumors, enables surgeons to excise the tumor margins completely, thereby reducing the risk of recurrence, and helps to make a more accurate prediction of the patient's prognosis than the conventional methods. AI also offers real-time training to neurosurgeons using virtual and 3D simulation, thereby increasing their confidence and skills during procedures. In addition, robotics is integrated into neurosurgery and identified to increase patient outcomes by making surgery less invasive. AI, including machine learning, is rigorously considered for its applications in the neurosurgical management of cerebrospinal tumors. This field requires further research focused on areas clinically essential in improving the outcome that is also economically feasible for clinical use. The authors suggest that data analysts and neurosurgeons collaborate to explore the full potential of AI.

Overall survival prediction for high-grade glioma patients using mathematical modeling of tumor cell infiltration

PURPOSE

This study aimed at applying a mathematical framework for the prediction of high-grade gliomas (HGGs) cell invasion into normal tissues for guiding the clinical target delineation, and at investigating the possibility of using tumor infiltration maps for patient overall survival (OS) prediction.

MATERIAL & METHODS

A model describing tumor infiltration into normal tissue was applied to 93 HGG cases. Tumor infiltration maps and corresponding isocontours with different cell densities were produced. ROC curves were used to seek correlations between the patient OS and the volume encompassed by a particular isocontour. Area-Under-the-Curve (AUC) values were used to determine the isocontour having the highest predictive ability. The optimal cut-off volume, having the highest sensitivity and specificity, for each isocontour was used to divide the patients in two groups for a Kaplan-Meier survival analysis.

RESULTS

The highest AUC value was obtained for the isocontour of cell densities 1000 cells/mm3 and 2000 cells/mm3, equal to 0.77 (p < 0.05). Correlation with the GTV yielded an AUC of 0.73 (p < 0.05). The Kaplan-Meier survival analysis using the 1000 cells/mm3 isocontour and the ROC optimal cut-off volume for patient group selection rendered a hazard ratio (HR) of 2.7 (p < 0.05), while the GTV rendered a HR = 1.6 (p < 0.05).

CONCLUSION

The simulated tumor cell invasion is a stronger predictor of overall survival than the segmented GTV, indicating the importance of using mathematical models for cell invasion to assist in the definition of the target for HGG patients.

Advances in artificial intelligence, robotics, augmented and virtual reality in neurosurgery

Neurosurgical practitioners undergo extensive and prolonged training to acquire diverse technical proficiencies, while neurosurgical procedures necessitate a substantial amount of pre-, post-, and intraoperative clinical data acquisition, making decisions, attention, and convalescence. The past decade witnessed an appreciable escalation in the significance of artificial intelligence (AI) in neurosurgery. AI holds significant potential in neurosurgery as it supplements the abilities of neurosurgeons to offer optimal interventional and non-interventional care to patients by improving prognostic and diagnostic outcomes in clinical therapy and assisting neurosurgeons in making decisions while surgical interventions to enhance patient outcomes. Other technologies including augmented reality, robotics, and virtual reality can assist and promote neurosurgical methods as well. Moreover, they play a significant role in generating, processing, as well as storing experimental and clinical data. Also, the usage of these technologies in neurosurgery is able to curtail the number of costs linked with surgical care and extend high-quality health care to a wider populace. This narrative review aims to integrate the results of articles that elucidate the role of the aforementioned technologies in neurosurgery.

Artificial Intelligence in Neurosurgery: A State-of-the-Art Review from Past to Future

In recent years, there has been a significant surge in discussions surrounding artificial intelligence (AI), along with a corresponding increase in its practical applications in various facets of everyday life, including the medical industry. Notably, even in the highly specialized realm of neurosurgery, AI has been utilized for differential diagnosis, pre-operative evaluation, and improving surgical precision. Many of these applications have begun to mitigate risks of intraoperative and postoperative complications and post-operative care. This article aims to present an overview of the principal published papers on the significant themes of tumor, spine, epilepsy, and vascular issues, wherein AI has been applied to assess its potential applications within neurosurgery. The method involved identifying high-cited seminal papers using PubMed and Google Scholar, conducting a comprehensive review of various study types, and summarizing machine learning applications to enhance understanding among clinicians for future utilization. Recent studies demonstrate that machine learning (ML) holds significant potential in neuro-oncological care, spine surgery, epilepsy management, and other neurosurgical applications. ML techniques have proven effective in tumor identification, surgical outcomes prediction, seizure outcome prediction, aneurysm prediction, and more, highlighting its broad impact and potential in improving patient management and outcomes in neurosurgery. This review will encompass the current state of research, as well as predictions for the future of AI within neurosurgery.

Predicting Survival in Patients with Brain Tumors: Current State-of-the-Art of AI Methods Applied to MRI

Given growing clinical needs, in recent years Artificial Intelligence (AI) techniques have increasingly been used to define the best approaches for survival assessment and prediction in patients with brain tumors. Advances in computational resources, and the collection of (mainly) public databases, have promoted this rapid development. This narrative review of the current state-of-the-art aimed to survey current applications of AI in predicting survival in patients with brain tumors, with a focus on Magnetic Resonance Imaging (MRI). An extensive search was performed on PubMed and Google Scholar using a Boolean research query based on MeSH terms and restricting the search to the period between 2012 and 2022. Fifty studies were selected, mainly based on Machine Learning (ML), Deep Learning (DL), radiomics-based methods, and methods that exploit traditional imaging techniques for survival assessment. In addition, we focused on two distinct tasks related to survival assessment: the first on the classification of subjects into survival classes (short and long-term or eventually short, mid and long-term) to stratify patients in distinct groups. The second focused on quantification, in days or months, of the individual survival interval. Our survey showed excellent state-of-the-art methods for the first, with accuracy up to ∼98%. The latter task appears to be the most challenging, but state-of-the-art techniques showed promising results, albeit with limitations, with C-Index up to ∼0.91. In conclusion, according to the specific task, the available computational methods perform differently, and the choice of the best one to use is non-univocal and dependent on many aspects. Unequivocally, the use of features derived from quantitative imaging has been shown to be advantageous for AI applications, including survival prediction. This evidence from the literature motivates further research in the field of AI-powered methods for survival prediction in patients with brain tumors, in particular, using the wealth of information provided by quantitative MRI techniques.

Artificial Intelligence for Survival Prediction in Brain Tumors on Neuroimaging

Survival prediction of patients affected by brain tumors provides essential information to guide surgical planning, adjuvant treatment selection, and patient counseling. Current reliance on clinical factors, such as Karnofsky Performance Status Scale, and simplistic radiological characteristics are, however, inadequate for survival prediction in tumors such as glioma that demonstrate molecular and clinical heterogeneity with variable survival outcomes. Advances in the domain of artificial intelligence have afforded powerful tools to capture a large number of hidden high-dimensional imaging features that reflect abundant information about tumor structure and physiology. Here, we provide an overview of current literature that apply computational analysis tools such as radiomics and machine learning methods to the pipeline of image preprocessing, tumor segmentation, feature extraction, and construction of classifiers to establish survival prediction models based on neuroimaging. We also discuss challenges relating to the development and evaluation of such models and explore ethical issues surrounding the future use of machine learning predictions.

A No-Math Primer on the Principles of Machine Learning for Radiologists

Machine learning is becoming increasingly important in both research and clinical applications in radiology due to recent technological developments, particularly in deep learning. As these technologies are translated toward clinical practice, there is a need for radiologists and radiology trainees to understand the basic principles behind them. This primer provides an accessible introduction to the vocabulary and concepts that are central to machine learning and relevant to the radiologist.

Preoperative Prediction of BRAF Mutation Status in Colorectal Cancer Using a Clinical-radiomics Model

RATIONALE AND OBJECTIVES

This study aimed to develop a clinically practical model to predict V-raf murine sarcoma viral oncogene homolog B1 (BRAF) mutation in colorectal cancer according to radiomic signatures based on computed tomography (CT) and clinical risk factors, and to determine the model's diagnostic accuracy for BRAF mutation status.

MATERIALS AND METHODS

This retrospective study included 140 patients with colorectal cancer. The significant clinical risk factors were used to build the clinical model; the least absolute shrinkage and selection operator algorithm was adopted to construct a radiomics signature according to imaging features of the tumor lesion, and stepwise logistic regression was applied to select the significant variables to develop the clinical-radiomics model. The predictive performance was evaluated by receiver operating characteristic curve analysis, calibration curve analysis, and decision curve analysis.

RESULTS

The radscore, generated by 5 selected radiomics features, demonstrated a favorable ability to predict BRAF mutation in both the training (area under the receiver operating characteristic curve [AUC] 0.93) and validation (AUC 0.87) cohorts. Subsequently, integrating two independent predictors (including the radscore and clinical risk factors) into a nomogram exhibited more favorable discriminatory performance, with the AUC improved to 0.95 and 0.88 in both cohorts. Moreover, the accuracy for predicting BRAF mutations was higher than that of the clinical model, ranging from 0.70 to 0.89.

CONCLUSION

The proposed CT-based radiomics signature is associated with BRAF mutations. The present study also proposes a combined model can potentially be applied in the individual preoperative prediction of BRAF mutation status in colorectal cancer.

ADVANCES IN KNOWLEDGE

CT-based radiomics showed satisfactory diagnostic significance for the BRAF status in colorectal cancer, the clinical-combined model may be applied in the individual preoperative prediction of BRAF mutation.

The role of artificial intelligence in paediatric neuroradiology

Imaging plays a fundamental role in the managing childhood neurologic, neurosurgical and neuro-oncological disease. Employing multi-parametric MRI techniques, such as spectroscopy and diffusion- and perfusion-weighted imaging, to the radiophenotyping of neuroradiologic conditions is becoming increasingly prevalent, particularly with radiogenomic analyses correlating imaging characteristics with molecular biomarkers of disease. However, integration into routine clinical practice remains elusive. With modern multi-parametric MRI now providing additional data beyond anatomy, informing on histology, biology and physiology, such metric-rich information can present as information overload to the treating radiologist and, as such, information relevant to an individual case can become lost. Artificial intelligence techniques are capable of modelling the vast radiologic, biological and clinical datasets that accompany childhood neurologic disease, such that this information can become incorporated in upfront prognostic modelling systems, with artificial intelligence techniques providing a plausible approach to this solution. This review examines machine learning approaches than can be used to underpin such artificial intelligence applications, with exemplars for each machine learning approach from the world literature. Then, within the specific use case of paediatric neuro-oncology, we examine the potential future contribution for such artificial intelligence machine learning techniques to offer solutions for patient care in the form of decision support systems, potentially enabling personalised medicine within this domain of paediatric radiologic practice.

Radiomics Analysis of Computed Tomography for Prediction of Thyroid Capsule Invasion in Papillary Thyroid Carcinoma: A Multi-Classifier and Two-Center Study

OBJECTIVE

To investigate the application of computed tomography (CT)-based radiomics model for prediction of thyroid capsule invasion (TCI) in papillary thyroid carcinoma (PTC).

METHODS

This retrospective study recruited 412 consecutive PTC patients from two independent institutions and randomly assigned to training (n=265), internal test (n=114) and external test (n=33) cohorts. Radiomics features were extracted from non-contrast (NC) and artery phase (AP) CT scans. We also calculated delta radiomics features, which are defined as the absolute differences between the extracted radiomics features. One-way analysis of variance and least absolute shrinkage and selection operator were used to select optimal radiomics features. Then, six supervised machine learning radiomics models (k-nearest neighbor, logistic regression, decision tree, linear support vector machine [L-SVM], Gaussian-SVM, and polynomial-SVM) were constructed. Univariate was used to select clinicoradiological risk factors. Combined models including optimal radiomics features and clinicoradiological risk factors were constructed by these six classifiers. The prediction performance was evaluated using the receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis (DCA).

RESULTS

In the internal test cohort, the best combined model (L-SVM, AUC=0.820 [95% CI 0.758-0.888]) performed better than the best radiomics model (L-SVM, AUC = 0.733 [95% CI 0.654-0.812]) and the clinical model (AUC = 0.709 [95% CI 0.649-0.783]). Combined-L-SVM model combines 23 radiomics features and 1 clinicoradiological risk factor (CT-reported TCI). In the external test cohort, the AUC was 0.776 (0.625-0.904) in the combined-L-SVM model, showing that the model is stable. DCA demonstrated that the combined model was clinically useful.

CONCLUSIONS

Our combined model based on machine learning incorporated with CT radiomics features and the clinicoradiological risk factor shows good predictive ability for TCI in PTC.

Automated Machine-Learning Framework Integrating Histopathological and Radiological Information for Predicting IDH1 Mutation Status in Glioma

Diffuse gliomas are the most common malignant primary brain tumors. Identification of isocitrate dehydrogenase 1 (IDH1) mutations aids the diagnostic classification of these tumors and the prediction of their clinical outcomes. While histology continues to play a key role in frozen section diagnosis, as a diagnostic reference and as a method for monitoring disease progression, recent research has demonstrated the ability of multi-parametric magnetic resonance imaging (MRI) sequences for predicting IDH genotypes. In this paper, we aim to improve the prediction accuracy of IDH1 genotypes by integrating multi-modal imaging information from digitized histopathological data derived from routine histological slide scans and the MRI sequences including T1-contrast (T1) and Fluid-attenuated inversion recovery imaging (T2-FLAIR). In this research, we have established an automated framework to process, analyze and integrate the histopathological and radiological information from high-resolution pathology slides and multi-sequence MRI scans. Our machine-learning framework comprehensively computed multi-level information including molecular level, cellular level, and texture level information to reflect predictive IDH genotypes. Firstly, an automated pre-processing was developed to select the regions of interest (ROIs) from pathology slides. Secondly, to interactively fuse the multimodal complementary information, comprehensive feature information was extracted from the pathology ROIs and segmented tumor regions (enhanced tumor, edema and non-enhanced tumor) from MRI sequences. Thirdly, a Random Forest (RF)-based algorithm was employed to identify and quantitatively characterize histopathological and radiological imaging origins, respectively. Finally, we integrated multi-modal imaging features with a machine-learning algorithm and tested the performance of the framework for IDH1 genotyping, we also provided visual and statistical explanation to support the understanding on prediction outcomes. The training and testing experiments on 217 pathologically verified IDH1 genotyped glioma cases from multi-resource validated that our fully automated machine-learning model predicted IDH1 genotypes with greater accuracy and reliability than models that were based on radiological imaging data only. The accuracy of IDH1 genotype prediction was 0.90 compared to 0.82 for radiomic result. Thus, the integration of multi-parametric imaging features for automated analysis of cross-modal biomedical data improved the prediction accuracy of glioma IDH1 genotypes.

Artificial Intelligence in Brain Tumour Surgery-An Emerging Paradigm

Artificial intelligence (AI) platforms have the potential to cause a paradigm shift in brain tumour surgery. Brain tumour surgery augmented with AI can result in safer and more effective treatment. In this review article, we explore the current and future role of AI in patients undergoing brain tumour surgery, including aiding diagnosis, optimising the surgical plan, providing support during the operation, and better predicting the prognosis. Finally, we discuss barriers to the successful clinical implementation, the ethical concerns, and we provide our perspective on how the field could be advanced.

Value of artificial intelligence model based on unenhanced computed tomography of urinary tract for preoperative prediction of calcium oxalate monohydrate stones in vivo

Background

Urolithiasis is a global disease with a high incidence and recurrence rate, and stone composition is closely related to the choice of treatment and preventive measures. Calcium oxalate monohydrate (COM) is the most common in clinical practice, which is hard and difficult to fragment. Preoperative identification of its components and selection of effective surgical methods can reduce the risk of patients having a second operation. Methods that can be used for stone composition analysis include infrared spectroscopy, X-ray diffraction, and polarized light microscopy, but they are all performed on stone specimens in vitro after surgery. This study aimed to design and develop an artificial intelligence (AI) model based on unenhanced computed tomography (CT) images of the urinary tract, and to investigate the predictive ability of the model for COM stones in vivo preoperatively, so as to provide surgeons with more accurate diagnostic information.

Methods

Preoperative unenhanced CT images of patients with urinary calculi whose components were determined by infrared spectroscopy in a single center were retrospectively analyzed, including 337 cases of COM stones and 170 of non-COM stones. All images were manually segmented and the image features were extracted, and randomly divided into the training and testing sets in a ratio of 7:3. The least absolute shrinkage and selection operation algorithm (LASSO) was used to construct the AI model, and classification of the training and testing sets was carried out.

Results

A total of 1,218 radiomics imaging features were extracted, and 8 features with non-zero coefficients were finally obtained. The sensitivity, specificity and accuracy of the AI model were 90.5%, 84.3% and 88.5% for the training set, and 90.1%, 84.3% and 88.3% for the testing set. The area under the curve was 0.935 for the training set and 0.933 for the testing set.

Conclusions

The AI model based on unenhanced CT images of the urinary tract can predict COM and non-COM stones in vivo preoperatively, and the model has high sensitivity, specificity and accuracy.

Neurosurgery and artificial intelligence

Neurosurgeons receive extensive and lengthy training to equip themselves with various technical skills, and neurosurgery require a great deal of pre-, intra- and postoperative clinical data collection, decision making, care and recovery. The last decade has seen a significant increase in the importance of artificial intelligence (AI) in neurosurgery. AI can provide a great promise in neurosurgery by complementing neurosurgeons' skills to provide the best possible interventional and noninterventional care for patients by enhancing diagnostic and prognostic outcomes in clinical treatment and help neurosurgeons with decision making during surgical interventions to improve patient outcomes. Furthermore, AI is playing a pivotal role in the production, processing and storage of clinical and experimental data. AI usage in neurosurgery can also reduce the costs associated with surgical care and provide high-quality healthcare to a broader population. Additionally, AI and neurosurgery can build a symbiotic relationship where AI helps to push the boundaries of neurosurgery, and neurosurgery can help AI to develop better and more robust algorithms. This review explores the role of AI in interventional and noninterventional aspects of neurosurgery during pre-, intra- and postoperative care, such as diagnosis, clinical decision making, surgical operation, prognosis, data acquisition, and research within the neurosurgical arena.

Combining MRI and Histologic Imaging Features for Predicting Overall Survival in Patients with Glioma

Purpose To test the hypothesis that combined features from MR and digital histopathologic images more accurately predict overall survival (OS) in patients with glioma compared with MRI or histopathologic features alone. Materials and Methods Multiparametric MR and histopathologic images in patients with a diagnosis of glioma (high- or low-grade glioma [HGG or LGG]) were obtained from The Cancer Imaging Archive (original images acquired 1983-2008). An extensive set of engineered features such as intensity, histogram, and texture were extracted from delineated tumor regions in MR and histopathologic images. Cox proportional hazard regression and support vector machine classification (SVC) models were applied to (a) MRI features only (MRI_cox/svc), histopathologic features only (HistoPath_cox/svc), and (c) combined MRI and histopathologic features (MRI+HistoPath_cox/svc) and evaluated in a split train-test configuration. Results A total of 171 patients (mean age, 51 years ± 15; 91 men) were included with HGG (n = 75) and LGG (n = 96). Median OS was 467 days (range, 3-4752 days) for all patients, 350 days (range, 15-1561 days) for HGG, and 595 days (range, 3-4752 days) for LGG. The MRI+HistoPath_cox model demonstrated higher concordance index (C-index) compared with MRI_cox and HistoPath_cox models on all patients (C-index, 0.79 vs 0.70 [P = .02; MRI_cox] and 0.67 [P = .01; HistoPath_cox]), patients with HGG (C-index, 0.78 vs 0.68 [P = .03; MRI_cox] and 0.64 [P = .01; HistoPath_cox]), and patients with LGG (C-index, 0.88 vs 0.62 [P = .008; MRI_cox] and 0.62 [P = .006; HistoPath_cox]). In binary classification, the MRI+HistoPath_svc model (area under the receiver operating characteristic curve [AUC], 0.86 [95% CI: 0.80, 0.95]) had higher performance than the MRI_svc model (AUC, 0.68 [95% CI: 0.50, 0.81]; P = .01) and the HistoPath_svc model (AUC, 0.72 [95% CI: 0.60, 0.85]; P = .04). Conclusion The model combining features from MR and histopathologic images had higher accuracy in predicting OS compared with the models with MR or histopathologic images alone. Keywords: Survival Prediction, Gliomas, Digital Pathology Imaging, MR Imaging, Machine Learning Supplemental material is available for this article.

A Fully Automated Deep Learning Network for Brain Tumor Segmentation

We developed a fully automated method for brain tumor segmentation using deep learning; 285 brain tumor cases with multiparametric magnetic resonance images from the BraTS2018 data set were used. We designed 3 separate 3D-Dense-UNets to simplify the complex multiclass segmentation problem into individual binary-segmentation problems for each subcomponent. We implemented a 3-fold cross-validation to generalize the network's performance. The mean cross-validation Dice-scores for whole tumor (WT), tumor core (TC), and enhancing tumor (ET) segmentations were 0.92, 0.84, and 0.80, respectively. We then retrained the individual binary-segmentation networks using 265 of the 285 cases, with 20 cases held-out for testing. We also tested the network on 46 cases from the BraTS2017 validation data set, 66 cases from the BraTS2018 validation data set, and 52 cases from an independent clinical data set. The average Dice-scores for WT, TC, and ET were 0.90, 0.84, and 0.80, respectively, on the 20 held-out testing cases. The average Dice-scores for WT, TC, and ET on the BraTS2017 validation data set, the BraTS2018 validation data set, and the clinical data set were as follows: 0.90, 0.80, and 0.78; 0.90, 0.82, and 0.80; and 0.85, 0.80, and 0.77, respectively. A fully automated deep learning method was developed to segment brain tumors into their subcomponents, which achieved high prediction accuracy on the BraTS data set and on the independent clinical data set. This method is promising for implementation into a clinical workflow.

Machine learning-based prediction of outcomes of the endoscopic endonasal approach in Cushing disease: is the future coming?

OBJECTIVE

Machine learning (ML) is an innovative method to analyze large and complex data sets. The aim of this study was to evaluate the use of ML to identify predictors of early postsurgical and long-term outcomes in patients treated for Cushing disease (CD).

METHODS

All consecutive patients in our center who underwent surgery for CD through the endoscopic endonasal approach were retrospectively reviewed. Study endpoints were gross-tumor removal (GTR), postsurgical remission, and long-term control of disease. Several demographic, radiological, and histological factors were assessed as potential predictors. For ML-based modeling, data were randomly divided into 2 sets with an 80% to 20% ratio for bootstrapped training and testing, respectively. Several algorithms were tested and tuned for the area under the curve (AUC).

RESULTS

The study included 151 patients. GTR was achieved in 137 patients (91%), and postsurgical hypersecretion remission was achieved in 133 patients (88%). At last follow-up, 116 patients (77%) were still in remission after surgery and in 21 patients (14%), CD was controlled with complementary treatment (overall, of 131 cases, 87% were under control at follow-up). At internal validation, the endpoints were predicted with AUCs of 0.81-1.00, accuracy of 81%-100%, and Brier scores of 0.035-0.151. Tumor size and invasiveness and histological confirmation of adrenocorticotropic hormone (ACTH)-secreting cells were the main predictors for the 3 endpoints of interest.

CONCLUSIONS

ML algorithms were used to train and internally validate robust models for all the endpoints, giving accurate outcome predictions in CD cases. This analytical method seems promising for potentially improving future patient care and counseling; however, careful clinical interpretation of the results remains necessary before any clinical adoption of ML. Moreover, further studies and increased sample sizes are definitely required before the widespread adoption of ML to the study of CD.

Machine learning applications to neuroimaging for glioma detection and classification: An artificial intelligence augmented systematic review

Glioma is the most common primary intraparenchymal tumor of the brain and the 5-year survival rate of high-grade glioma is poor. Magnetic resonance imaging (MRI) is essential for detecting, characterizing and monitoring brain tumors but definitive diagnosis still relies on surgical pathology. Machine learning has been applied to the analysis of MRI data in glioma research and has the potential to change clinical practice and improve patient outcomes. This systematic review synthesizes and analyzes the current state of machine learning applications to glioma MRI data and explores the use of machine learning for systematic review automation. Various datapoints were extracted from the 153 studies that met inclusion criteria and analyzed. Natural language processing (NLP) analysis involved keyword extraction, topic modeling and document classification. Machine learning has been applied to tumor grading and diagnosis, tumor segmentation, non-invasive genomic biomarker identification, detection of progression and patient survival prediction. Model performance was generally strong (AUC = 0.87 ± 0.09; sensitivity = 0.87 ± 0.10; specificity = 0.0.86 ± 0.10; precision = 0.88 ± 0.11). Convolutional neural network, support vector machine and random forest algorithms were top performers. Deep learning document classifiers yielded acceptable performance (mean 5-fold cross-validation AUC = 0.71). Machine learning tools and data resources were synthesized and summarized to facilitate future research. Machine learning has been widely applied to the processing of MRI data in glioma research and has demonstrated substantial utility. NLP and transfer learning resources enabled the successful development of a replicable method for automating the systematic review article screening process, which has potential for shortening the time from discovery to clinical application in medicine.

Prognostic prediction of hypertensive intracerebral hemorrhage using CT radiomics and machine learning

OBJECTIVES

Spontaneous intracerebral hemorrhage remains a major cause of death and disability throughout the world. We tried to establish accurate long-term outcome prediction models for hypertensive intracerebral hemorrhage (HICH) using CT radiomics and machine learning.

METHODS

In a retrospective study of 270 patients with HICH between June 2013 and June 2018, CT images and patients' 6-month outcome based on the modified Rankin Scale were collected. Hematomas on CT images were selected as volumes of interests (VOIs), and 1,029 radiomics features of the VOIs were extracted. Based on correlations with patients' outcome, radiomics features underwent dimensionality reduction analyses. Then, the support vector machine (SVM), k-nearest neighbor (KNN), logistic regression (LR), decision tree (DT), random forest (RF), and XGBoost algorithms were applied with the screened features to establish prognostic prediction models of HICH. Accuracies of all models were compared.

RESULTS

Eighteen radiomics features were screened as prognosis-associated radiomics signature of HICH based on the variance threshold, SelectKBest, and least absolute shrinkage and selection operator (LASSO) regression models. Patients were randomly allocated into training (n = 215) and validation (n = 55) sets. Accuracies of all 6 machine learning algorithms in the validation set exceeded 80%. The sensitivity, specificity, and accuracy in the validation set were 93.3%, 92.5%, and 92.7% for the RF model and 92.3%, 88.1%, and 89.1% for the XGBoost model, respectively, which were the best two among all models.

CONCLUSIONS

Taking advantage of radiomics and machine learning, we established accurate prognostic prediction models of HICH. The RF model and XGBoost model returned the best accuracies.

Differentiation Between Ependymoma and Medulloblastoma in Children with Radiomics Approach

RATIONALE AND OBJECTIVES

Ependymoma (EP) and medulloblastoma (MB) of children are similar in age, location, manifestations and symptoms. Therefore, it is difficult to differentiate them through visual observation in clinical diagnosis. The aim of this study is to investigate the effectiveness of radiomics and machine-learning techniques on multimodal magnetic resonance imaging (MRI) in distinguish EP from MB.

MATERIALS AND METHODS

Three dimensional (3D) tumors were semi-automatic segmented by radiologists from postcontrast T1-weighted images and apparent diffusion coefficient maps in 51 patients (24 EPs, 27 MBs). Then, we extracted radiomics features and further reduced them by three feature selection methods. For each feature selection method, 4 classifiers were adopted which yield 12 different models. After extensive crossvalidation, pairwise test were carried out in receiver operating characteristic curves to explore performance of these models.

RESULTS

The radiomics model built with multivariable logistic regression as feature selection method and random forests as classifier had the best performance, area under the curve achieved 0.91 (95 % confidence interval 0.787-0.968). Five relevant features were highly correlated to discriminate EP and MB, which may used as imaging biomarkers to predict the kinds of tumors.

CONCLUSION

The combination of radiomics and machine-learning approach on 3D multimodal MRI could well distinguish EP and MB of childhood, which assistant doctors in clinical diagnosis. Since there is no uniform model to obtained best performance for every specific data set, it is necessary to try different combination methods.

The Nomogram of MRI-based Radiomics with Complementary Visual Features by Machine Learning Improves Stratification of Glioblastoma Patients: A Multicenter Study

BACKGROUND

Glioblastomas (GBMs) represent both the most common and the most highly malignant primary brain tumors. The subjective visual imaging features from MRI make it challenging to predict the overall survival (OS) of GBM. Radiomics can quantify image features objectively as an emerging technique. A pragmatic and objective method in the clinic to assess OS is strongly in need.

PURPOSE

To construct a radiomics nomogram to stratify GBM patients into long- vs. short-term survival.

STUDY TYPE

Retrospective.

POPULATION

One-hundred and fifty-eight GBM patients from Brain Tumor Segmentation Challenge 2018 (BRATS2018) were for model construction and 32 GBM patients from the local hospital for external validation.

FIELD STRENGTH/SEQUENCE

1.5 T and 3.0 T MRI Scanners, T₁ WI, T₂ WI, T₂ FLAIR, and contrast-enhanced T₁ WI sequences ASSESSMENT: All patients were divided into long-term or short-term based on a survival of greater or fewer than 12 months. All BRATS2018 subjects were divided into training and test sets, and images were assessed for ependymal and pia mater involvement (EPI) and multifocality by three experienced neuroradiologists. All tumor tissues from multiparametric MRI were fully automatically segmented into three subregions to calculate the radiomic features. Based on the training set, the most powerful radiomic features were selected to constitute radiomic signature.

STATISTICAL TESTS

Receiver operating characteristic (ROC) curve, sensitivity, specificity, and the Hosmer-Lemeshow test.

RESULTS

The nomogram had a survival prediction accuracy of 0.878 and 0.875, a specificity of 0.875 and 0.583, and a sensitivity of 0.704 and 0.833, respectively, in the training and test set. The ROC curve showed the accuracy of the nomogram, radiomic signature, age, and EPI for external validation set were 0.858, 0.826, 0.664, and 0.66 in the validate set, respectively.

DATA CONCLUSION

Radiomics nomogram integrated with radiomic signature, EPI, and age was found to be robust for the stratification of GBM patients into long- vs. short-term survival.

LEVEL OF EVIDENCE

3 TECHNICAL EFFICACY STAGE: 2.

A neuro evolutionary algorithm for patient calibrated prediction of survival in Glioblastoma patients

BACKGROUND AND OBJECTIVES

Glioblastoma multiforme (GBM) is the most common and malignant type of primary brain tumors. Radiation therapy (RT) plus concomitant and adjuvant Temozolomide (TMZ) constitute standard treatment of GBM. Existing models for GBM growth do not consider the effect of different schedules on tumor growth and patient survival. However, clinical trials show that treatment schedule and drug dosage significantly affect patient survival. The goal is to provide a patient calibrated model for predicting survival according to the treatment schedule.

METHODS

We propose a top-down method based on artificial neural networks (ANN) and genetic algorithm (GA) to predict survival of GBM patients. A feed forward undercomplete Autoencoder network is integrated with the neuro-evolutionary (NE) algorithm in order to extract a compressed representation of input clinical data. The proposed NE algorithm uses GA to obtain optimal architecture of a multi-layer perceptron (MLP). Taguchi L₁₆ orthogonal design of experiments is used to tune parameters of the proposed NE algorithm. Finally, the optimal MLP is used to predict survival of GBM patients.

RESULTS

Data from 8 related clinical trials have been collected and integrated to train the model. From 847 evaluable cases, 719 were used for train and validation and the remaining 128 cases were used to test the model. Mean absolute error of the predictions on the test data is 0.087 months which shows excellent performance of the proposed model in predicting survival of the patients. Also, the results show that the proposed NE algorithm is superior to other existing models in both the mean and variability of the prediction error.

Promises and Perils of Artificial Intelligence in Neurosurgery

Artificial intelligence (AI)-facilitated clinical automation is expected to become increasingly prevalent in the near future. AI techniques may permit rapid and detailed analysis of the large quantities of clinical data generated in modern healthcare settings, at a level that is otherwise impossible by humans. Subsequently, AI may enhance clinical practice by pushing the limits of diagnostics, clinical decision making, and prognostication. Moreover, if combined with surgical robotics and other surgical adjuncts such as image guidance, AI may find its way into the operating room and permit more accurate interventions, with fewer errors. Despite the considerable hype surrounding the impending medical AI revolution, little has been written about potential downsides to increasing clinical automation. These may include both direct and indirect consequences. Directly, faulty, inadequately trained, or poorly understood algorithms may produce erroneous results, which may have wide-scale impact. Indirectly, increasing use of automation may exacerbate de-skilling of human physicians due to over-reliance, poor understanding, overconfidence, and lack of necessary vigilance of an automated clinical workflow. Many of these negative phenomena have already been witnessed in other industries that have already undergone, or are undergoing "automation revolutions," namely commercial aviation and the automotive industry. This narrative review explores the potential benefits and consequences of the anticipated medical AI revolution from a neurosurgical perspective.

Artificial intelligence and hybrid imaging: the best match for personalized medicine in oncology

Artificial intelligence (AI) refers to a field of computer science aimed to perform tasks typically requiring human intelligence. Currently, AI is recognized in the broader technology radar within the five key technologies which emerge for their wide-ranging applications and impact in communities, companies, business, and value chain framework alike. However, AI in medical imaging is at an early phase of development, and there are still hurdles to take related to reliability, user confidence, and adoption. The present narrative review aimed to provide an overview on AI-based approaches (distributed learning, statistical learning, computer-aided diagnosis and detection systems, fully automated image analysis tool, natural language processing) in oncological hybrid medical imaging with respect to clinical tasks (detection, contouring and segmentation, prediction of histology and tumor stage, prediction of mutational status and molecular therapies targets, prediction of treatment response, and outcome). Particularly, AI-based approaches have been briefly described according to their purpose and, finally lung cancer-being one of the most extensively malignancy studied by hybrid medical imaging-has been used as illustrative scenario. Finally, we discussed clinical challenges and open issues including ethics, validation strategies, effective data-sharing methods, regulatory hurdles, educational resources, and strategy to facilitate the interaction among different stakeholders. Some of the major changes in medical imaging will come from the application of AI to workflow and protocols, eventually resulting in improved patient management and quality of life. Overall, several time-consuming tasks could be automatized. Machine learning algorithms and neural networks will permit sophisticated analysis resulting not only in major improvements in disease characterization through imaging, but also in the integration of multiple-omics data (i.e., derived from pathology, genomic, proteomics, and demographics) for multi-dimensional disease featuring. Nevertheless, to accelerate the transition of the theory to practice a sustainable development plan considering the multi-dimensional interactions between professionals, technology, industry, markets, policy, culture, and civil society directed by a mindset which will allow talents to thrive is necessary.

Machine learning in neurosurgery: a global survey

Background

Recent technological advances have led to the development and implementation of machine learning (ML) in various disciplines, including neurosurgery. Our goal was to conduct a comprehensive survey of neurosurgeons to assess the acceptance of and attitudes toward ML in neurosurgical practice and to identify factors associated with its use.

Methods

The online survey consisted of nine or ten mandatory questions and was distributed in February and March 2019 through the European Association of Neurosurgical Societies (EANS) and the Congress of Neurosurgeons (CNS).

Results

Out of 7280 neurosurgeons who received the survey, we received 362 responses, with a response rate of 5%, mainly in Europe and North America. In total, 103 neurosurgeons (28.5%) reported using ML in their clinical practice, and 31.1% in research. Adoption rates of ML were relatively evenly distributed, with 25.6% for North America, 30.9% for Europe, 33.3% for Latin America and the Middle East, 44.4% for Asia and Pacific and 100% for Africa with only two responses. No predictors of clinical ML use were identified, although academic settings and subspecialties neuro-oncology, functional, trauma and epilepsy predicted use of ML in research. The most common applications were for predicting outcomes and complications, as well as interpretation of imaging.

Conclusions

This report provides a global overview of the neurosurgical applications of ML. A relevant proportion of the surveyed neurosurgeons reported clinical experience with ML algorithms. Future studies should aim to clarify the role and potential benefits of ML in neurosurgery and to reconcile these potential advantages with bioethical considerations.

Diagnostic utility of Restriction Spectrum Imaging in the characterization of the peritumoral brain zone in glioblastoma: Analysis of overall and progression-free survival

PURPOSE

We studied the ability of Restriction Spectrum Imaging (RSI), a novel advanced diffusion imaging technique, to estimate levels of cellularity in different glioblastoma regions, evaluated their prognostic value compared with established clinical diffusion metrics such as fractional anisotropy (FA) and mean diffusivity (MD).

METHODS

Forty-two patients with untreated glioblastoma, IDH-wildtype, were examined with an advanced MRI tumor protocol. The region of interest (ROI) was obtained from the contrast-enhancing part of tumor and the peritumoral brain zones and then co-registered with RSI-cellularity index, FA and MD maps. Histogram parameters of diffusion metrics were assessed for all ROI locations and compared to MGMT promoter methylation status and survival. The ability of RSI-cellularity index, FA, and MD to stratify survival and were assessed by Cox proportional hazard regression, adjusted for significant clinical predictors.

RESULTS

The highest RSI-cellularity index was measured in contrast-enhancing tumor core with a negative gradient from tumor core to the periphery of peritumoral zone with predictive accuracy 81 % (P < 0.001). Shorter overall survival was significant associated with higher RSI-cellularity index (hazard ratio (HR) 3.6, 95 % confidence interval (CI) 1.3-9.5, P = 0.002) with synchronal decrease in MD (HR 0.31, 95 %CI 0.1-0.8, P = 0.008) in the contrast-enhanced tumor core. This association was also consistent for RSI-cellularity index value measured in the peri-enhancing zone (HR 3.6, 95 % CI 1.0-12.3, P = 0.041). No statistically significant differences were noted between RSI-cellularity index, FA, nor MD and MGMT promoter methylation.

CONCLUSION

RSI-cellularity index may be used as prognostic biomarker to improve risk stratification in patients with glioblastoma.

Is Artificial Intelligence Better Than Human Clinicians in Predicting Patient Outcomes?

In contrast with medical imaging diagnostics powered by artificial intelligence (AI), in which deep learning has led to breakthroughs in recent years, patient outcome prediction poses an inherently challenging problem because it focuses on events that have not yet occurred. Interestingly, the performance of machine learning-based patient outcome prediction models has rarely been compared with that of human clinicians in the literature. Human intuition and insight may be sources of underused predictive information that AI will not be able to identify in electronic data. Both human and AI predictions should be investigated together with the aim of achieving a human-AI symbiosis that synergistically and complementarily combines AI with the predictive abilities of clinicians.

Multi-Parametric Deep Learning Model for Prediction of Overall Survival after Postoperative Concurrent Chemoradiotherapy in Glioblastoma Patients

This study aimed to investigate the performance of a deep learning-based survival-prediction model, which predicts the overall survival (OS) time of glioblastoma patients who have received surgery followed by concurrent chemoradiotherapy (CCRT). The medical records of glioblastoma patients who had received surgery and CCRT between January 2011 and December 2017 were retrospectively reviewed. Based on our inclusion criteria, 118 patients were selected and semi-randomly allocated to training and test datasets (3:1 ratio, respectively). A convolutional neural network–based deep learning model was trained with magnetic resonance imaging (MRI) data and clinical profiles to predict OS. The MRI was reconstructed by using four pulse sequences (22 slices) and nine images were selected based on the longest slice of glioblastoma by a physician for each pulse sequence. The clinical profiles consist of personal, genetic, and treatment factors. The concordance index (C-index) and integrated area under the curve (iAUC) of the time-dependent area-under-the-curve curves of each model were calculated to evaluate the performance of the survival-prediction models. The model that incorporated clinical and radiomic features showed a higher C-index (0.768 (95% confidence interval (CI): 0.759, 0.776)) and iAUC (0.790 (95% CI: 0.783, 0.797)) than the model using clinical features alone (C-index = 0.693 (95% CI: 0.685, 0.701); iAUC = 0.723 (95% CI: 0.716, 0.731)) and the model using radiomic features alone (C-index = 0.590 (95% CI: 0.579, 0.600); iAUC = 0.614 (95% CI: 0.607, 0.621)). These improvements to the C-indexes and iAUCs were validated using the 1000-times bootstrapping method; all were statistically significant (p < 0.001). This study suggests the synergistic benefits of using both clinical and radiomic parameters. Furthermore, it indicates the potential of multi-parametric deep learning models for the survival prediction of glioblastoma patients.

Machine learning for predicting long-term kidney allograft survival: a scoping review

Supervised machine learning (ML) is a class of algorithms that "learn" from existing input-output pairs, which is gaining popularity in pattern recognition for classification and prediction problems. In this scoping review, we examined the use of supervised ML algorithms for the prediction of long-term allograft survival in kidney transplant recipients. Data sources included PubMed, the Cumulative Index to Nursing and Allied Health Literature, and the Institute for Electrical and Electronics Engineers (IEEE) Xplore libraries from inception to November 2019. We screened titles and abstracts and potentially eligible full-text reports to select studies and subsequently abstracted the data. Eleven studies were identified. Decision trees were the most commonly used method (n = 8), followed by artificial neural networks (ANN) (n = 4) and Bayesian belief networks (n = 2). The area under receiver operating curve (AUC) was the most common measure of discrimination (n = 7), followed by sensitivity (n = 5) and specificity (n = 4). Model calibration examining the reliability in risk prediction was performed using either the Pearson r or the Hosmer-Lemeshow test in four studies. One study showed that logistic regression had comparable performance to ANN, while another study demonstrated that ANN performed better in terms of sensitivity, specificity, and accuracy, as compared with a Cox proportional hazards model. We synthesized the evidence related to the comparison of ML techniques with traditional statistical approaches for prediction of long-term allograft survival in patients with a kidney transplant. The methodological and reporting quality of included studies was poor. Our study also demonstrated mixed results in terms of the predictive potential of the models.

Precision Digital Oncology: Emerging Role of Radiomics-based Biomarkers and Artificial Intelligence for Advanced Imaging and Characterization of Brain Tumors

Advances in computerized image analysis and the use of artificial intelligence-based approaches for image-based analysis and construction of prediction algorithms represent a new era for noninvasive biomarker discovery. In recent literature, it has become apparent that radiologic images can serve as mineable databases that contain large amounts of quantitative features with potential clinical significance. Extraction and analysis of these quantitative features is commonly referred to as texture or radiomic analysis. Numerous studies have demonstrated applications for texture and radiomic characterization methods for assessing brain tumors to improve noninvasive predictions of tumor histologic characteristics, molecular profile, distinction of treatment-related changes, and prediction of patient survival. In this review, the current use and future potential of texture or radiomic-based approaches with machine learning for brain tumor image analysis and prediction algorithm construction will be discussed. This technology has the potential to advance the value of diagnostic imaging by extracting currently unused information on medical scans that enables more precise, personalized therapy; however, significant barriers must be overcome if this technology is to be successfully implemented on a wide scale for routine use in the clinical setting. Keywords: Adults and Pediatrics, Brain/Brain Stem, CNS, Computer Aided Diagnosis (CAD), Computer Applications-General (Informatics), Image Postprocessing, Informatics, Neural Networks, Neuro-Oncology, Oncology, Treatment Effects, Tumor Response Supplemental material is available for this article. © RSNA, 2020.

Machine learning assisted DSC-MRI radiomics as a tool for glioma classification by grade and mutation status

Background

Combining MRI techniques with machine learning methodology is rapidly gaining attention as a promising method for staging of brain gliomas. This study assesses the diagnostic value of such a framework applied to dynamic susceptibility contrast (DSC)-MRI in classifying treatment-naïve gliomas from a multi-center patients into WHO grades II-IV and across their isocitrate dehydrogenase (IDH) mutation status.

Methods

Three hundred thirty-three patients from 6 tertiary centres, diagnosed histologically and molecularly with primary gliomas (IDH-mutant = 151 or IDH-wildtype = 182) were retrospectively identified. Raw DSC-MRI data was post-processed for normalised leakage-corrected relative cerebral blood volume (rCBV) maps. Shape, intensity distribution (histogram) and rotational invariant Haralick texture features over the tumour mask were extracted. Differences in extracted features across glioma grades and mutation status were tested using the Wilcoxon two-sample test. A random-forest algorithm was employed (2-fold cross-validation, 250 repeats) to predict grades or mutation status using the extracted features.

Results

Shape, distribution and texture features showed significant differences across mutation status. WHO grade II-III differentiation was mostly driven by shape features while texture and intensity feature were more relevant for the III-IV separation. Increased number of features became significant when differentiating grades further apart from one another. Gliomas were correctly stratified by mutation status in 71% and by grade in 53% of the cases (87% of the gliomas grades predicted with distance less than 1).

Conclusions

Despite large heterogeneity in the multi-center dataset, machine learning assisted DSC-MRI radiomics hold potential to address the inherent variability and presents a promising approach for non-invasive glioma molecular subtyping and grading.

Predicting Inpatient Length of Stay After Brain Tumor Surgery: Developing Machine Learning Ensembles to Improve Predictive Performance

BACKGROUND

Current outcomes prediction tools are largely based on and limited by regression methods. Utilization of machine learning (ML) methods that can handle multiple diverse inputs could strengthen predictive abilities and improve patient outcomes. Inpatient length of stay (LOS) is one such outcome that serves as a surrogate for patient disease severity and resource utilization.

OBJECTIVE

To develop a novel method to systematically rank, select, and combine ML algorithms to build a model that predicts LOS following craniotomy for brain tumor.

METHODS

A training dataset of 41 222 patients who underwent craniotomy for brain tumor was created from the National Inpatient Sample. Twenty-nine ML algorithms were trained on 26 preoperative variables to predict LOS. Trained algorithms were ranked by calculating the root mean square logarithmic error (RMSLE) and top performing algorithms combined to form an ensemble. The ensemble was externally validated using a dataset of 4592 patients from the National Surgical Quality Improvement Program. Additional analyses identified variables that most strongly influence the ensemble model predictions.

RESULTS

The ensemble model predicted LOS with RMSLE of .555 (95% confidence interval, .553-.557) on internal validation and .631 on external validation. Nonelective surgery, preoperative pneumonia, sodium abnormality, or weight loss, and non-White race were the strongest predictors of increased LOS.

CONCLUSION

An ML ensemble model predicts LOS with good performance on internal and external validation, and yields clinical insights that may potentially improve patient outcomes. This systematic ML method can be applied to a broad range of clinical problems to improve patient care.

Improved Prediction of Surgical Resectability in Patients with Glioblastoma using an Artificial Neural Network

In managing a patient with glioblastoma (GBM), a surgeon must carefully consider whether sufficient tumour can be removed so that the patient can enjoy the benefits of decompression and cytoreduction, without impacting on the patient’s neurological status. In a previous study we identified the five most important anatomical features on a pre-operative MRI that are predictive of surgical resectability and used them to develop a simple, objective, and reproducible grading system. The objective of this study was to apply an artificial neural network (ANN) to improve the prediction of surgical resectability in patients with GBM. Prospectively maintained databases were searched to identify adult patients with supratentorial GBM that underwent craniotomy and resection. Performance of the ANN was evaluated against logistic regression and the standard grading system by analysing their Receiver Operator Characteristic (ROC) curves; Area Under Curve (AUC) and accuracy were calculated and compared using Wilcoxon signed rank test with a value of p < 0.05 considered statistically significant. In all, 135 patients were included, of which 33 (24.4%) were found to have complete excision of all contrast-enhancing tumour. The AUC and accuracy were significantly greater using the ANN compared to the standard grading system (0.87 vs. 0.79 and 83% vs. 80% respectively; p < 0.01 in both cases). In conclusion, an ANN allows for the improved prediction of surgical resectability in patients with GBM.

Emerging MRI Techniques to Redefine Treatment Response in Patients With Glioblastoma

Glioblastoma is the most common and most malignant primary brain tumor. Despite aggressive multimodal treatment, its prognosis remains poor. Even with continuous developments in MRI, which has provided us with newer insights into the diagnosis and understanding of tumor biology, response assessment in the posttherapy setting remains challenging. We believe that the integration of additional information from advanced neuroimaging techniques can further improve the diagnostic accuracy of conventional MRI. In this article, we review the utility of advanced neuroimaging techniques such as diffusion-weighted imaging, diffusion tensor imaging, perfusion-weighted imaging, proton magnetic resonance spectroscopy, and chemical exchange saturation transfer in characterizing and evaluating treatment response in patients with glioblastoma. We will also discuss the existing challenges and limitations of using these techniques in clinical settings and possible solutions to avoiding pitfalls in study design, data acquisition, and analysis for future studies. LEVEL OF EVIDENCE: 2 TECHNICAL EFFICACY STAGE: 3 J. Magn. Reson. Imaging 2020;52:978-997.

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

We applied machine learning algorithms for differentiation of posterior fossa tumors using apparent diffusion coefficient (ADC) histogram analysis and structural MRI findings. A total of 256 patients with intra-axial posterior fossa tumors were identified, of whom 248 were included in machine learning analysis, with at least 6 representative subjects per each tumor pathology. The ADC histograms of solid components of tumors, structural MRI findings, and patients' age were applied to construct decision models using Classification and Regression Tree analysis. We also compared different machine learning classification algorithms (i.e., naïve Bayes, random forest, neural networks, support vector machine with linear and polynomial kernel) for dichotomized differentiation of the 5 most common tumors in our cohort: metastasis (n = 65), hemangioblastoma (n = 44), pilocytic astrocytoma (n = 43), ependymoma (n = 27), and medulloblastoma (n = 26). The decision tree model could differentiate seven tumor histopathologies with terminal nodes yielding up to 90% accurate classification rates. In receiver operating characteristics (ROC) analysis, the decision tree model achieved greater area under the curve (AUC) for differentiation of pilocytic astrocytoma (p = 0.020); and atypical teratoid/rhabdoid tumor ATRT (p = 0.001) from other types of neoplasms compared to the official clinical report. However, neuroradiologists' interpretations had greater accuracy in differentiating metastases (p = 0.001). Among different machine learning algorithms, random forest models yielded the highest accuracy in dichotomized classification of the 5 most common tumor types; and in multiclass differentiation of all tumor types random forest yielded an averaged AUC of 0.961 in training datasets, and 0.873 in validation samples. Our study demonstrates the potential application of machine learning algorithms and decision trees for accurate differentiation of brain tumors based on pretreatment MRI. Using easy to apply and understandable imaging metrics, the proposed decision tree model can help radiologists with differentiation of posterior fossa tumors, especially in tumors with similar qualitative imaging characteristics. In particular, our decision tree model provided more accurate differentiation of pilocytic astrocytomas from ATRT than by neuroradiologists in clinical reads.

Development and validation of a MRI-based radiomics signature for prediction of KRAS mutation in rectal cancer

OBJECTIVE

To develop a T2-weighted (T2W) image-based radiomics signature for the individual prediction of KRAS mutation status in patients with rectal cancer.

METHODS

Three hundred four consecutive patients from center I with pathologically diagnosed rectal adenocarcinoma (training dataset, n = 213; internal validation dataset, n = 91) were enrolled in our retrospective study. The patients from center II (n = 86) were selected as an external validation dataset. A total of 960 imaging features were extracted from high-resolution T2W images for each patient. Five steps, mainly univariate statistical tests, were applied for feature selection. Subsequently, three classification methods, i.e., logistic regression (LR), decision tree (DT), and support vector machine (SVM) algorithm, were applied to develop the radiomics signature for KRAS prediction in the training dataset. The predictive performance was evaluated by receiver operating characteristics curve (ROC) analysis, calibration curve, and decision curve analysis (DCA).

RESULTS

Seven radiomics features were screened as a KRAS-associated radiomics signature of rectal cancer. Our best prediction model was obtained with SVM classifiers with AUC of 0.722 (95%CI, 0.654-0.790) in the training dataset. This was validated in the internal and external validation datasets with good calibration, and the corresponding AUCs were 0.682 (95% CI, 0.569-0.794) and 0.714 (95% CI, 0.602-0.827), respectively. DCA confirmed its clinical usefulness.

CONCLUSIONS

The proposed T2WI-based radiomics signature has a moderate performance to predict KRAS status, and may be useful for supplementing genomic analysis to determine KRAS expression in rectal cancer patients.

KEY POINTS

• T2WI-based radiomics showed a moderate diagnostic significance for KRAS status. • The best prediction model was obtained with SVM classifier. • The baseline clinical and histopathological characteristics were not associated with KRAS mutation.

Association of bone mineral density with bone texture attributes extracted using routine magnetic resonance imaging

OBJECTIVE

Dual-energy X-ray absorptiometry (DXA)-derived bone mineral density (BMD) often fails to predict fragility fractures. Quantitative textural analysis using magnetic resonance imaging (MRI) may potentially yield useful radiomic features to predict fractures. We aimed to investigate the correlation between BMD and texture attributes (TAs) extracted from MRI scans and the interobserver reproducibility of the analysis.

METHODS

Forty-nine volunteers underwent lumbar spine 1.5-T MRI and DXA. Three-dimensional (3-D) gray-level co-occurrence matrices were measured from routine sagittal T2 fast spin-echo images using the IBEX software. Twenty-two TAs were extracted from 3-D segmented L3 vertebrae. The estimated concordance coefficient was calculated using linear regression analysis. A Pearson correlation coefficient analysis was performed to evaluate the correlation between BMD and the TAs. Interobserver reproducibility was assessed with the concordance coefficient described by Lin.

RESULTS

The results revealed a fair-to-moderate significant correlation between BMD and 13 TAs (r=-0.20 to 0.39; p<0.05). Eight TAs (autocorrelation, energy, homogeneity 1, homogeneity 1.1, maximum probability, sum average, sum variance, and inverse difference normalized) negatively correlated with BMD (r=-0.20 to -0.38; p<0.05), whereas five TAs (dissimilarity, difference entropy, entropy, sum entropy, and information measure corr 1) positively correlated with BMD (r=0.29-0.39; p<0.05). The interobserver agreement was almost perfect for all significant TAs (95% confidence interval, 0.92-1.00; p<0.05).

CONCLUSION

Specific TAs could be reliably extracted from routine MRI and correlated with BMD. Our results encourage future evaluation of the potential usefulness of quantitative texture measurements from MRI scans for predicting fragility fractures.

The Impact of MRI Features and Observer Confidence on the Treatment Decision-Making for Patients with Untreated Glioma

In a blind, dual-center, multi-observer setting, we here identify the pre-treatment radiologic features by Magnetic Resonance Imaging (MRI) associated with subsequent treatment options in patients with glioma. Study included 220 previously untreated adult patients from two institutions (94 + 126 patients) with a histopathologically confirmed diagnosis of glioma after surgery. Using a blind, cross-institutional and randomized setup, four expert neuroradiologists recorded radiologic features, suggested glioma grade and corresponding confidence. The radiologic features were scored using the Visually AcceSAble Rembrandt Images (VASARI) standard. Results were retrospectively compared to patient treatment outcomes. Our findings show that patients receiving a biopsy or a subtotal resection were more likely to have a tumor with pathological MRI-signal (by T2-weighted Fluid-Attenuated Inversion Recovery) crossing the midline (Hazard Ratio; HR = 1.30 [1.21–1.87], P < 0.001), and those receiving a biopsy sampling more often had multifocal lesions (HR = 1.30 [1.16–1.64], P < 0.001). For low-grade gliomas (N = 50), low observer confidence in the radiographic readings was associated with less chance of a total resection (P = 0.002) and correlated with the use of a more comprehensive adjuvant treatment protocol (Spearman = 0.48, P < 0.001). This study may serve as a guide to the treating physician by identifying the key radiologic determinants most likely to influence the treatment decision-making process.

Development and Validation of a MRI-Based Radiomics Prognostic Classifier in Patients with Primary Glioblastoma Multiforme

RATIONALE AND OBJECTIVES

Glioblastoma multiforme (GBM) is the most common and deadly type of primary malignant tumor of the central nervous system. Accurate risk stratification is vital for a more personalized approach in GBM management. The purpose of this study is to develop and validate a MRI-based prognostic quantitative radiomics classifier in patients with newly diagnosed GBM and to evaluate whether the classifier allows stratification with improved accuracy over the clinical and qualitative imaging features risk models.

METHODS

Clinical and MR imaging data of 127 GBM patients were obtained from the Cancer Genome Atlas and the Cancer Imaging Archive. Regions of interest corresponding to high signal intensity portions of tumor were drawn on postcontrast T1-weighted imaging (post-T1WI) on the 127 patients (allocated in a 2:1 ratio into a training [n = 85] or validation [n = 42] set), then 3824 radiomics features per patient were extracted. The dimension of these radiomics features were reduced using the minimum redundancy maximum relevance algorithm, then Cox proportional hazard regression model was used to build a radiomics classifier for predicting overall survival (OS). The value of the radiomics classifier beyond clinical (gender, age, Karnofsky performance status, radiation therapy, chemotherapy, and type of resection) and VASARI features for OS was assessed with multivariate Cox proportional hazards model. Time-dependent receiver operating characteristic curve analysis was used to assess the predictive accuracy.

RESULTS

A classifier using four post-T1WI-MRI radiomics features built on the training dataset could successfully separate GBM patients into low- or high-risk group with a significantly different OS in training (HR, 6.307 [95% CI, 3.475-11.446]; p < 0.001) and validation set (HR, 3.646 [95% CI, 1.709-7.779]; p < 0.001). The area under receiver operating characteristic curve of radiomics classifier (training, 0.799; validation, 0.815 for 12-month) was higher compared to that of the clinical risk model (Karnofsky performance status, radiation therapy; training, 0.749; validation, 0.670 for 12-month), and none of the qualitative imaging features was associated with OS. The predictive accuracy was further improved when combined the radiomics classifier with clinical data (training, 0.819; validation: 0.851 for 12-month).

CONCLUSION

A classifier using radiomics features allows preoperative prediction of survival and risk stratification of patients with GBM, and it shows improved performance compared to that of clinical and qualitative imaging features models.

Natural and Artificial Intelligence in Neurosurgery: A Systematic Review

BACKGROUND

Machine learning (ML) is a domain of artificial intelligence that allows computer algorithms to learn from experience without being explicitly programmed.

OBJECTIVE

To summarize neurosurgical applications of ML where it has been compared to clinical expertise, here referred to as "natural intelligence."

METHODS

A systematic search was performed in the PubMed and Embase databases as of August 2016 to review all studies comparing the performance of various ML approaches with that of clinical experts in neurosurgical literature.

RESULTS

Twenty-three studies were identified that used ML algorithms for diagnosis, presurgical planning, or outcome prediction in neurosurgical patients. Compared to clinical experts, ML models demonstrated a median absolute improvement in accuracy and area under the receiver operating curve of 13% (interquartile range 4-21%) and 0.14 (interquartile range 0.07-0.21), respectively. In 29 (58%) of the 50 outcome measures for which a P-value was provided or calculated, ML models outperformed clinical experts (P < .05). In 18 of 50 (36%), no difference was seen between ML and expert performance (P > .05), while in 3 of 50 (6%) clinical experts outperformed ML models (P < .05). All 4 studies that compared clinicians assisted by ML models vs clinicians alone demonstrated a better performance in the first group.

CONCLUSION

We conclude that ML models have the potential to augment the decision-making capacity of clinicians in neurosurgical applications; however, significant hurdles remain associated with creating, validating, and deploying ML models in the clinical setting. Shifting from the preconceptions of a human-vs-machine to a human-and-machine paradigm could be essential to overcome these hurdles.

Machine learning in predicting graft failure following kidney transplantation: A systematic review of published predictive models

INTRODUCTION

Machine learning has been increasingly used to develop predictive models to diagnose different disease conditions. The heterogeneity of the kidney transplant population makes predicting graft outcomes extremely challenging. Several kidney graft outcome prediction models have been developed using machine learning, and are available in the literature. However, a systematic review of machine learning based prediction methods applied to kidney transplant has not been done to date. The main aim of our study was to perform an in-depth systematic analysis of different machine learning methods used to predict graft outcomes among kidney transplant patients, and assess their usefulness as an aid to decision-making.

METHODS

A systemic review of machine learning methods used to predict graft outcomes among kidney transplant patients was carried out using a search of the Medline, the Cumulative Index to Nursing and Allied Health Literature, EMBASE, PsycINFO and Cochrane databases.

RESULTS

A total of 295 articles were identified and extracted. Of these, 18 met the inclusion criteria. Most of the studies were published in the United States after 2010. The population size used to develop the models varied from 80 to 92,844, and the number of features in the models ranged from 6 to 71. The most common machine learning methods used were artificial neural networks, decision trees and Bayesian belief networks. Most of the machine learning based predictive models predicted graft failure with high sensitivity and specificity. Only one machine learning based prediction model had modelled time-to-event (survival) information. Seven studies compared the predictive performance of machine learning models with traditional regression methods and the performance of machine learning methods was found to be mixed, when compared with traditional regression methods.

CONCLUSION

There was a wide variation in the size of the study population and the input variables used. However, the prediction accuracy provided mixed results when machine learning and traditional predictive methods are compared. Based on reported gains in predictive performance, machine learning has the potential to improve kidney transplant outcome prediction and aid medical decision making.