Machine Learning in Differentiating Gliomas from Primary CNS Lymphomas: A Systematic Review, Reporting Quality, and Risk of Bias Assessment

Inter and intra-operator reliability of Lekholm and Zarb classification and proposal of a novel radiomic data-driven clustering for qualitative assessment of edentulous alveolar ridges

OBJECTIVES

The present study was conducted to evaluate the reproducibility of Lekholm and Zarb classification system (L&Z) for bone quality assessment of edentulous alveolar ridges and to investigate the potential of a data-driven approach for bone quality classification.

MATERIALS AND METHODS

Twenty-six expert clinicians were asked to classify 110 CBCT cross-sections according to L&Z classification (T0). The same evaluation was repeated after one month with the images put in a different order (T1). Intra- and inter-examiner agreement analyses were performed using Cohen's kappa coefficient (CK) and Fleiss' kappa coefficient (FK), respectively. Additionally, radiomic features extraction was performed from 3D edentulous ridge blocks derived from the same 110 CBCTs, and unsupervised clustering using 3 different clustering methods was used to identify patterns in the obtained data.

RESULTS

Intra-examiner agreement between T0 and T1 was weak (CK 0.515). Inter-examiner agreement at both time points was minimal (FK at T0: 0.273; FK at T1: 0.243). The three different unsupervised clustering methods based on radiomic features aggregated the 110 CBCTs in three groups in the same way.

CONCLUSIONS

The results showed low agreement among clinicians when using L&Z classification, indicating that the system may not be as reliable as previously thought. The present study suggests the possible application of a reproducible data-driven approach based on radiomics for the classification of edentulous alveolar ridges, with potential implications for improving clinical outcomes. Further research is needed to determine the clinical significance of these findings and to develop more standardized and accurate methods for assessing bone quality of edentulous alveolar ridges.

How does deep learning/machine learning perform in comparison to radiologists in distinguishing glioblastomas (or grade IV astrocytomas) from primary CNS lymphomas?: a meta-analysis and systematic review

BACKGROUND

Several studies have been published comparing deep learning (DL)/machine learning (ML) to radiologists in differentiating PCNSLs from GBMs with equivocal results. We aimed to perform this meta-analysis to evaluate the diagnostic accuracy of ML/DL versus radiologists in classifying PCNSL versus GBM using MRI.

METHODOLOGY

The study was performed in accordance with PRISMA guidelines. Data was extracted and interpreted by two researchers with 12 and 23 years' experience, respectively, and QUADAS-2 tool was used for quality and risk-bias assessment. We constructed contingency tables to derive sensitivity, specificity accuracy, summary receiver operating characteristic (SROC) curve, and the area under the curve (AUC).

RESULTS

Our search identified 11 studies, of which 8 satisfied our inclusion criteria and restricted the analysis in each study to reporting the model showing highest accuracy, with a total sample size of 1159 patients. The random effects model showed a pooled sensitivity of 0.89 [95% CI:0.84-0.92] for ML and 0.82 [95% CI:0.76-0.87] for radiologists. Pooled specificity was 0.88 [95% CI: 0.84-0.91] for ML and 0.90 [95% CI: 0.81-0.95] for radiologists. Pooled accuracy was 0.88 [95% CI: 0.86-0.90] for ML and 0.86 [95% CI: 0.78-0.91] for radiologists. Pooled AUC of ML was 0.94 [95% CI:0.92-0.96]and for radiologists, it was 0.90 [95% CI: 0.84-0.93].

CONCLUSIONS

MRI-based ML/DL techniques can complement radiologists to improve the accuracy of classifying GBMs from PCNSL, possibly reduce the need for a biopsy, and avoid any unwanted neurosurgical resection of a PCNSL.

Advancements in Image-Based Models for High-Grade Gliomas Might Be Accelerated

The first half of 2022 saw the publication of several major research advances in image-based models and artificial intelligence applications to optimize treatment strategies for high-grade gliomas, the deadliest brain tumors. We review them and discuss the barriers that delay their entry into clinical practice; particularly, the small sample size and the heterogeneity of the study designs and methodologies used. We will also write about the poor and late palliation that patients suffering from high-grade glioma can count on at the end of life, as well as the current legislative instruments, with particular reference to Italy. We suggest measures to accelerate the gradual progress in image-based models and end of life care for patients with high-grade glioma.

Deep learning aided preoperative diagnosis of primary central nervous system lymphoma

The preoperative distinction between glioblastoma (GBM) and primary central nervous system lymphoma (PCNSL) can be difficult, even for experts, but is highly relevant. We aimed to develop an easy-to-use algorithm, based on a convolutional neural network (CNN) to preoperatively discern PCNSL from GBM and systematically compare its performance to experienced neurosurgeons and radiologists. To this end, a CNN-based on DenseNet169 was trained with the magnetic resonance (MR)-imaging data of 68 PCNSL and 69 GBM patients and its performance compared to six trained experts on an external test set of 10 PCNSL and 10 GBM. Our neural network predicted PCNSL with an accuracy of 80% and a negative predictive value (NPV) of 0.8, exceeding the accuracy achieved by clinicians (73%, NPV 0.77). Combining expert rating with automated diagnosis in those cases where experts dissented yielded an accuracy of 95%. Our approach has the potential to significantly augment the preoperative radiological diagnosis of PCNSL.

Systematic Literature Review of Machine Learning Algorithms Using Pretherapy Radiologic Imaging for Glioma Molecular Subtype Prediction

BACKGROUND

The molecular profile of gliomas is a prognostic indicator for survival, driving clinical decision-making for treatment. Pathology-based molecular diagnosis is challenging because of the invasiveness of the procedure, exclusion from neoadjuvant therapy options, and the heterogeneous nature of the tumor.

PURPOSE

We performed a systematic review of algorithms that predict molecular subtypes of gliomas from MR Imaging.

DATA SOURCES

Data sources were Ovid Embase, Ovid MEDLINE, Cochrane Central Register of Controlled Trials, Web of Science.

STUDY SELECTION

Per the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, 12,318 abstracts were screened and 1323 underwent full-text review, with 85 articles meeting the inclusion criteria.

DATA ANALYSIS

We compared prediction results from different machine learning approaches for predicting molecular subtypes of gliomas. Bias analysis was conducted for each study, following the Prediction model Risk Of Bias Assessment Tool (PROBAST) guidelines.

DATA SYNTHESIS

Isocitrate dehydrogenase mutation status was reported with an area under the curve and accuracy of 0.88 and 85% in internal validation and 0.86 and 87% in limited external validation data sets, respectively. For the prediction of O6-methylguanine-DNA methyltransferase promoter methylation, the area under the curve and accuracy in internal validation data sets were 0.79 and 77%, and in limited external validation, 0.89 and 83%, respectively. PROBAST scoring demonstrated high bias in all articles.

LIMITATIONS

The low number of external validation and studies with incomplete data resulted in unequal data analysis. Comparing the best prediction pipelines of each study may introduce bias.

CONCLUSIONS

While the high area under the curve and accuracy for the prediction of molecular subtypes of gliomas are reported in internal and external validation data sets, limited use of external validation and the increased risk of bias in all articles may present obstacles for clinical translation of these techniques.

Imaging Cancer in Neuroradiology

Neuroimaging plays a pivotal role in the diagnosis, management, and prognostication of brain tumors. Recently, the World Health Organization published the fifth edition of the WHO Classification of Tumors of the Central Nervous System (CNS5), which places greater emphasis on tumor genetics and molecular markers to complement the existing histological and immunohistochemical approaches. Recent advances in computational power allowed modern neuro-oncological imaging to move from a strictly morphology-based discipline to advanced neuroimaging techniques with quantifiable tissue characteristics such as tumor cellularity, microstructural organization, hemodynamic, functional, and metabolic features, providing more precise tumor diagnosis and management. The aim of this review is to highlight the key imaging features of the recently published CNS5, outlining the current imaging standards and summarizing the latest advances in neuro-oncological imaging techniques and their role in complementing traditional brain tumor imaging and management.

Clinical implementation of artificial intelligence in neuroradiology with development of a novel workflow-efficient picture archiving and communication system-based automated brain tumor segmentation and radiomic feature extraction

Purpose

Personalized interpretation of medical images is critical for optimum patient care, but current tools available to physicians to perform quantitative analysis of patient’s medical images in real time are significantly limited. In this work, we describe a novel platform within PACS for volumetric analysis of images and thus development of large expert annotated datasets in parallel with radiologist performing the reading that are critically needed for development of clinically meaningful AI algorithms. Specifically, we implemented a deep learning-based algorithm for automated brain tumor segmentation and radiomics extraction, and embedded it into PACS to accelerate a supervised, end-to- end workflow for image annotation and radiomic feature extraction.

Materials and methods

An algorithm was trained to segment whole primary brain tumors on FLAIR images from multi-institutional glioma BraTS 2021 dataset. Algorithm was validated using internal dataset from Yale New Haven Health (YHHH) and compared (by Dice similarity coefficient [DSC]) to radiologist manual segmentation. A UNETR deep-learning was embedded into Visage 7 (Visage Imaging, Inc., San Diego, CA, United States) diagnostic workstation. The automatically segmented brain tumor was pliable for manual modification. PyRadiomics (Harvard Medical School, Boston, MA) was natively embedded into Visage 7 for feature extraction from the brain tumor segmentations.

Results

UNETR brain tumor segmentation took on average 4 s and the median DSC was 86%, which is similar to published literature but lower than the RSNA ASNR MICCAI BRATS challenge 2021. Finally, extraction of 106 radiomic features within PACS took on average 5.8 ± 0.01 s. The extracted radiomic features did not vary over time of extraction or whether they were extracted within PACS or outside of PACS. The ability to perform segmentation and feature extraction before radiologist opens the study was made available in the workflow. Opening the study in PACS, allows the radiologists to verify the segmentation and thus annotate the study.

Conclusion

Integration of image processing algorithms for tumor auto-segmentation and feature extraction into PACS allows curation of large datasets of annotated medical images and can accelerate translation of research into development of personalized medicine applications in the clinic. The ability to use familiar clinical tools to revise the AI segmentations and natively embedding the segmentation and radiomic feature extraction tools on the diagnostic workstation accelerates the process to generate ground-truth data.

Classifying primary central nervous system lymphoma from glioblastoma using deep learning and radiomics based machine learning approach - a systematic review and meta-analysis

Background

Glioblastoma (GBM) and primary central nervous system lymphoma (PCNSL) are common in elderly yet difficult to differentiate on MRI. Their management and prognosis are quite different. Recent surge of interest in predictive analytics, using machine learning (ML) from radiomic features and deep learning (DL) for diagnosing, predicting response and prognosticating disease has evinced interest among radiologists and clinicians. The objective of this systematic review and meta-analysis was to evaluate the deep learning & ML algorithms in classifying PCNSL from GBM.

Methods

The authors performed a systematic review of the literature from MEDLINE, EMBASE and the Cochrane central trials register for the search strategy in accordance with PRISMA guidelines to select and evaluate studies that included themes of ML, DL, AI, GBM, PCNSL. All studies reporting on ML algorithms or DL that for differentiating PCNSL from GBM on MR imaging were included. These studies were further narrowed down to focus on works published between 2018 and 2021. Two researchers independently conducted the literature screening, database extraction and risk bias assessment. The extracted data was synthesised and analysed by forest plots. Outcomes assessed were test characteristics such as accuracy, sensitivity, specificity and balanced accuracy.

Results

Ten articles meeting the eligibility criteria were identified addressing use of ML and DL in training and validation classifiers to distinguish PCNSL from GBM on MR imaging. The total sample size was 1311 in the included studies. ML approach was used in 6 studies while DL in 4 studies. The lowest reported sensitivity was 80%, while the highest reported sensitivity was 99% in studies in which ML and DL was directly compared with the gold standard histopathology. The lowest reported specificity was 87% while the highest reported specificity was 100%. The highest reported balanced accuracy was 100% and the lowest was 84%.

Conclusions

Extensive search of the database revealed a limited number of studies that have applied ML or DL to differentiate PCNSL from GBM. Of the currently published studies, Both DL & ML algorithms have demonstrated encouraging results and certainly have the potential to aid neurooncologists in taking preoperative decisions in the future leading to not only reduction in morbidities but also be cost effective.

Role of Positron Emission Tomography in Primary Central Nervous System Lymphoma

Simple Summary

Primary central nervous system lymphoma (PCNSL) is a rare but highly aggressive lymphoma with increasing incidence in immunocompetent patients. To date, the only established biomarkers for survival are age and functional status. Currently, the magnetic resonance imaging (MRI) criteria of the International Collaborative Group on Primary Central Nervous System Lymphoma are the only ones recommended for follow-up. However, early occurrence of recurrence after treatment in patients with a complete response on MRI raises the question of its performance in assessing residual disease. While the use of ¹⁸F-fluorodeoxyglucose body positron emission tomography for identification of systemic disease has been established and can be pivotal in patient treatment decisions, the role of brain PET scan is less clear. Here we review the potential role of PET in the management of patients with PCNSL, both at diagnosis and for follow-up under treatment.

Abstract

The incidence of primary central nervous system lymphoma has increased over the past two decades in immunocompetent patients and the prognosis remains poor. A diagnosis and complete evaluation of the patient is needed without delay, but histologic evaluation is not always available and PCNSL can mimic a variety of brain lesions on MRI. In this article, we review the potential role of ¹⁸F-FDG PET for the diagnosis of PCNSL in immunocompetent and immunocompromised patients. Its contribution to systemic assessment at the time of diagnosis has been well established by expert societies over the past decade. In addition, ¹⁸F-FDG provides valuable information for differential diagnosis and outcome prediction. The literature also shows the potential role of ¹⁸F-FDG as a therapeutic evaluation tool during the treatment and the end of the treatment. Finally, we present several new radiotracers that may have a potential role in the management of PCNSL in the future.

Machine Learning Tools for Image-Based Glioma Grading and the Quality of Their Reporting: Challenges and Opportunities

Technological innovation has enabled the development of machine learning (ML) tools that aim to improve the practice of radiologists. In the last decade, ML applications to neuro-oncology have expanded significantly, with the pre-operative prediction of glioma grade using medical imaging as a specific area of interest. We introduce the subject of ML models for glioma grade prediction by remarking upon the models reported in the literature as well as by describing their characteristic developmental workflow and widely used classifier algorithms. The challenges facing these models-including data sources, external validation, and glioma grade classification methods -are highlighted. We also discuss the quality of how these models are reported, explore the present and future of reporting guidelines and risk of bias tools, and provide suggestions for the reporting of prospective works. Finally, this review offers insights into next steps that the field of ML glioma grade prediction can take to facilitate clinical implementation.