Diffusion and perfusion MRI radiomics obtained from deep learning segmentation provides reproducible and comparable diagnostic model to human in post-treatment glioblastoma

Uncovering hidden treasures: Mapping morphological changes in the differentiation of human mesenchymal stem cells to osteoblasts using deep learning

Deep Learning (DL) is becoming an increasingly popular technology being employed in life sciences research due to its ability to perform complex and time-consuming tasks with significantly greater speed, accuracy, and reproducibility than human researchers - allowing them to dedicate their time to more complex tasks. One potential application of DL is to analyze cell images taken by microscopes. Quantitative analysis of cell microscopy images remain a challenge - with manual cell characterization requiring excessive amounts of time and effort. DL can address these issues, by quickly extracting such data and enabling rigorous, empirical analysis of images. Here, DL is used to quantitively analyze images of Mesenchymal Stem Cells (MSCs) differentiating into Osteoblasts (OBs), tracking morphological changes throughout this transition. The changes in morphology throughout the differentiation protocol provide evidence for a distinct path of morphological transformations that the cells undergo in their transition, with changes in perimeter being observable before changes in eceentricity. Subsequent differentiation experiments can be quantitatively compared with our dataset to concretely evaluate how different conditions affect differentiation and this paper can also be used as a guide for researchers on how to utilize DL workflows in their own labs.

Achieving imaging and computational reproducibility on multiparametric MRI radiomics features in brain tumor diagnosis: phantom and clinical validation

OBJECTIVES

The Image Biomarker Standardization Initiative has helped improve the computational reproducibility of MRI radiomics features. Nonetheless, the MRI sequences and features with high imaging reproducibility are yet to be established. To determine reproducible multiparametric MRI radiomics features across test-retest, multi-scanner, and computational reproducibility comparisons, and to evaluate their clinical value in brain tumor diagnosis.

METHODS

To assess reproducibility, T1-weighted imaging (T1WI), T2-weighted imaging (T2WI), and diffusion-weighted imaging (DWI) were acquired from three 3-T MRI scanners using standardized phantom, and radiomics features were extracted using two computational algorithms. Reproducible radiomics features were selected when the concordance correlation coefficient value above 0.9 across multiple sessions, scanners, and computational algorithms. Random forest classifiers were trained with reproducible features (n = 117) and validated in a clinical cohort (n = 50) to evaluate whether features with high reproducibility improved the differentiation of glioblastoma from primary central nervous system lymphomas (PCNSLs).

RESULTS

Radiomics features from T2WI demonstrated higher repeatability (65-94%) than those from DWI (38-48%) or T1WI (2-92%). Across test-retest, multi-scanner, and computational comparisons, T2WI provided 41 reproducible features, DWI provided six, and T1WI provided two. The performance of the classification model with reproducible features was higher than that using non-reproducible features in both training set (AUC, 0.916 vs. 0.877) and validation set (AUC, 0.957 vs. 0.869).

CONCLUSION

Radiomics features with high reproducibility across multiple sessions, scanners, and computational algorithms were identified, and they showed higher diagnostic performance than non-reproducible radiomics features in the differentiation of glioblastoma from PCNSL.

CLINICAL RELEVANCE STATEMENT

By identifying the radiomics features showing higher multi-machine reproducibility, our results also demonstrated higher radiomics diagnostic performance in the differentiation of glioblastoma from PCNSL, paving the way for further research designs and clinical application in neuro-oncology.

KEY POINTS

• Highly reproducible radiomics features across multiple sessions, scanners, and computational algorithms were identified using phantom and applied to clinical diagnosis. • Radiomics features from T2-weighted imaging were more reproducible than those from T1-weighted and diffusion-weighted imaging. • Radiomics features with good reproducibility had better diagnostic performance for brain tumors than features with poor reproducibility.

Post-treatment imaging of gliomas: challenging the existing dogmas

Gliomas are the commonest malignant central nervous system tumours in adults and imaging is the cornerstone of diagnosis, treatment, and post-treatment follow-up of these patients. With the ever-evolving treatment strategies post-treatment imaging and interpretation in glioma remains challenging, more so with the advent of anti-angiogenic drugs and immunotherapy, which can significantly alter the appearance in this setting, thus making interpretation of routine imaging findings such as contrast enhancement, oedema, and mass effect difficult to interpret. This review details the various methods of management of glioma including the upcoming novel therapies and their impact on imaging findings, with a comprehensive description of the imaging findings in conventional and advanced imaging techniques. A systematic appraisal for the existing and emerging techniques of imaging in these settings and their clinical application including various response assessment guidelines and artificial intelligence based response assessment will also be discussed.

A subregion-based RadioFusionOmics model discriminates between grade 4 astrocytoma and glioblastoma on multisequence MRI

PURPOSE

To explore a subregion-based RadioFusionOmics (RFO) model for discrimination between adult-type grade 4 astrocytoma and glioblastoma according to the 2021 WHO CNS5 classification.

METHODS

329 patients (40 grade 4 astrocytomas and 289 glioblastomas) with histologic diagnosis was retrospectively collected from our local institution and The Cancer Imaging Archive (TCIA). The volumes of interests (VOIs) were obtained from four multiparametric MRI sequences (T1WI, T1WI + C, T2WI, T2-FLAIR) using (1) manual segmentation of the non-enhanced tumor (nET), enhanced tumor (ET), and peritumoral edema (pTE), and (2) K-means clustering of four habitats (H1: high T1WI + C, high T2-FLAIR; (2) H2: high T1WI + C, low T2-FLAIR; (3) H3: low T1WI + C, high T2-FLAIR; and (4) H4: low T1WI + C, low T2-FLAIR). The optimal VOI and best MRI sequence combination were determined. The performance of the RFO model was evaluated using the area under the precision-recall curve (AUPRC) and the best signatures were identified.

RESULTS

The two best VOIs were manual VOI3 (putative peritumoral edema) and clustering H34 (low T1WI + C, high T2-FLAIR (H3) combined with low T1WI + C and low T2-FLAIR (H4)). Features fused from four MRI sequences ([Formula: see text]) outperformed those from either a single sequence or other sequence combinations. The RFO model that was trained using fused features [Formula: see text] achieved the AUPRC of 0.972 (VOI3) and 0.976 (H34) in the primary cohort (p = 0.905), and 0.971 (VOI3) and 0.974 (H34) in the testing cohort (p = 0.402).

CONCLUSION

The performance of subregions defined by clustering was comparable to that of subregions that were manually defined. Fusion of features from the edematous subregions of multiple MRI sequences by the RFO model resulted in differentiation between grade 4 astrocytoma and glioblastoma.

Convolutional Neural Networks for Glioma Segmentation and Prognosis: A Systematic Review

Deep learning (DL) is poised to redefine the way medical images are processed and analyzed. Convolutional neural networks (CNNs), a specific type of DL architecture, are exceptional for high-throughput processing, allowing for the effective extraction of relevant diagnostic patterns from large volumes of complex visual data. This technology has garnered substantial interest in the field of neuro-oncology as a promising tool to enhance medical imaging throughput and analysis. A multitude of methods harnessing MRI-based CNNs have been proposed for brain tumor segmentation, classification, and prognosis prediction. They are often applied to gliomas, the most common primary brain cancer, to classify subtypes with the goal of guiding therapy decisions. Additionally, the difficulty of repeating brain biopsies to evaluate treatment response in the setting of often confusing imaging findings provides a unique niche for CNNs to help distinguish the treatment response to gliomas. For example, glioblastoma, the most aggressive type of brain cancer, can grow due to poor treatment response, can appear to grow acutely due to treatment-related inflammation as the tumor dies (pseudo-progression), or falsely appear to be regrowing after treatment as a result of brain damage from radiation (radiation necrosis). CNNs are being applied to separate this diagnostic dilemma. This review provides a detailed synthesis of recent DL methods and applications for intratumor segmentation, glioma classification, and prognosis prediction. Furthermore, this review discusses the future direction of MRI-based CNN in the field of neuro-oncology and challenges in model interpretability, data availability, and computation efficiency.

The application value of deep learning in the background of precision medicine in glioblastoma

Introduction: Glioblastoma is a highly malignant central nervous system tumor, World Health Organization Ⅳ, glioblastoma is the most common primary malignancy, due to its own specificity and complexity, different patients often benefit from the current conventional treatment regimen because of different molecular subtypes, in the context of precision medicine, the application of deep learning to identify the salient features of tumors on brain imaging, prognostic predictive assessment combined with clinical data to maximize the benefits of each patient from the treatment regimen is a non-invasive and feasible regimen. Methods: We conducted a comprehensive review of the existing literature on the role of deep learning in glioblastomas, covering molecular classification and diagnosis, prognosis assessment. Results: Data based on a variety of magnetic resonance imaging sequences, genetic information, and clinical combinations enable noninvasive predictive tumor diagnosis of glioblastoma and assess overall survival and treatment response accuracy. For images, standardized image acquisition and data extraction techniques can be effectively translated into learning models for clinical practice. However, it must be recognized that interventions in the treatment of glioblastoma using deep learning are still in their infancy, and the robustness of the model is challenged, as the current total number of glioblastoma samples is insufficient for large-scale experimental methods, which is directly related to the difficulty of application of the model. Conclusion: Compared to radiomics and shallow machine learning, deep learning can be a more robust, non-invasive, and effective approach, providing more valuable information as clinicians develop personalized medical protocols for glioblastoma patients.

Radiomics characterization of tissues in an animal brain tumor model imaged using dynamic contrast enhanced (DCE) MRI

Here, we investigate radiomics-based characterization of tumor vascular and microenvironmental properties in an orthotopic rat brain tumor model measured using dynamic-contrast-enhanced (DCE) MRI. Thirty-two immune compromised-RNU rats implanted with human U-251N cancer cells were imaged using DCE-MRI (7Tesla, Dual-Gradient-Echo). The aim was to perform pharmacokinetic analysis using a nested model (NM) selection technique to classify brain regions according to vasculature properties considered as the source of truth. A two-dimensional convolutional-based radiomics analysis was performed on the raw-DCE-MRI of the rat brains to generate dynamic radiomics maps. The raw-DCE-MRI and respective radiomics maps were used to build 28 unsupervised Kohonen self-organizing-maps (K-SOMs). A Silhouette-Coefficient (SC), k-fold Nested-Cross-Validation (k-fold-NCV), and feature engineering analyses were performed on the K-SOMs' feature spaces to quantify the distinction power of radiomics features compared to raw-DCE-MRI for classification of different Nested Models. Results showed that eight radiomics features outperformed respective raw-DCE-MRI in prediction of the three nested models. The average percent difference in SCs between radiomics features and raw-DCE-MRI was: 29.875% ± 12.922%, p < 0.001. This work establishes an important first step toward spatiotemporal characterization of brain regions using radiomics signatures, which is fundamental toward staging of tumors and evaluation of tumor response to different treatments.

Immune checkpoint inhibition: a future guided by radiology

The limitation of the function of antitumour immune cells is a common hallmark of cancers that enables their survival. As such, the potential of immune checkpoint inhibition (ICI) acts as a paradigm shift in the treatment of a range of cancers but has not yet been fully capitalised. Combining minimally and non-invasive locoregional therapies offered by radiologists with ICI is now an active field of research with the aim of furthering therapeutic capabilities in medical oncology. In parallel to this impending advancement, the "imaging toolbox" available to radiologists is also growing, enabling more refined tumour characterisation as well as greater accuracy in evaluating responses to therapy. Options range from metabolite labelling to cellular localisation to immune checkpoint screening. It is foreseeable that these novel imaging techniques will be integrated into personalised treatment algorithms. This growth in the field must include updating the current standardised imaging criteria to ensure they are fit for purpose. Such criteria is crucial to both appropriately guide clinical decision-making regarding next steps of treatment, but also provide reliable prognosis. Quantitative approaches to these novel imaging techniques are also already being investigated to further optimise personalised therapeutic decision-making. The therapeutic potential of specific ICIs and locoregional therapies could be determined before administration thus limiting unnecessary side-effects whilst maintaining efficacy. Several radiological aspects of oncological care are advancing simultaneously. Therefore, it is essential that each development is assessed for clinical use and optimised to ensure the best treatment decisions are being offered to the patient. In this review, we discuss state of the art advances in novel functional imaging techniques in the field of immuno-oncology both pre-clinically and clinically.

Bibliometric research on the developments of artificial intelligence in radiomics toward nervous system diseases

Background

The growing interest suggests that the widespread application of radiomics has facilitated the development of neurological disease diagnosis, prognosis, and classification. The application of artificial intelligence methods in radiomics has increasingly achieved outstanding prediction results in recent years. However, there are few studies that have systematically analyzed this field through bibliometrics. Our destination is to study the visual relationships of publications to identify the trends and hotspots in radiomics research and encourage more researchers to participate in radiomics studies.

Methods

Publications in radiomics in the field of neurological disease research can be retrieved from the Web of Science Core Collection. Analysis of relevant countries, institutions, journals, authors, keywords, and references is conducted using Microsoft Excel 2019, VOSviewer, and CiteSpace V. We analyze the research status and hot trends through burst detection.

Results

On October 23, 2022, 746 records of studies on the application of radiomics in the diagnosis of neurological disorders were retrieved and published from 2011 to 2023. Approximately half of them were written by scholars in the United States, and most were published in Frontiers in Oncology, European Radiology, Cancer, and SCIENTIFIC REPORTS. Although China ranks first in the number of publications, the United States is the driving force in the field and enjoys a good academic reputation. NORBERT GALLDIKS and JIE TIAN published the most relevant articles, while GILLIES RJ was cited the most. RADIOLOGY is a representative and influential journal in the field. "Glioma" is a current attractive research hotspot. Keywords such as "machine learning," "brain metastasis," and "gene mutations" have recently appeared at the research frontier.

Conclusion

Most of the studies focus on clinical trial outcomes, such as the diagnosis, prediction, and prognosis of neurological disorders. The radiomics biomarkers and multi-omics studies of neurological disorders may soon become a hot topic and should be closely monitored, particularly the relationship between tumor-related non-invasive imaging biomarkers and the intrinsic micro-environment of tumors.

Emerging role of artificial intelligence in diagnosis, classification and clinical management of glioma

Glioma represents a dominant primary intracranial malignancy in the central nervous system. Artificial intelligence that mainly includes machine learning, and deep learning computational approaches, presents a unique opportunity to enhance clinical management of glioma through improving tumor segmentation, diagnosis, differentiation, grading, treatment, prediction of clinical outcomes (prognosis, and recurrence), molecular features, clinical classification, characterization of the tumor microenvironment, and drug discovery. A growing body of recent studies apply artificial intelligence-based models to disparate data sources of glioma, covering imaging modalities, digital pathology, high-throughput multi-omics data (especially emerging single-cell RNA sequencing and spatial transcriptome), etc. While these early findings are promising, future studies are required to normalize artificial intelligence-based models to improve the generalizability and interpretability of the results. Despite prominent issues, targeted clinical application of artificial intelligence approaches in glioma will facilitate the development of precision medicine of this field. If these challenges can be overcome, artificial intelligence has the potential to profoundly change the way patients with or at risk of glioma are provided with more rational care.

Combination of ultrafast dynamic contrast-enhanced MRI-based radiomics and artificial neural network in assessing BI-RADS 4 breast lesions: Potential to avoid unnecessary biopsies

Objectives

To investigate whether combining radiomics extracted from ultrafast dynamic contrast-enhanced MRI (DCE-MRI) with an artificial neural network enables differentiation of MR BI-RADS 4 breast lesions and thereby avoids false-positive biopsies.

Methods

This retrospective study consecutively included patients with MR BI-RADS 4 lesions. The ultrafast imaging was performed using Differential sub-sampling with cartesian ordering (DISCO) technique and the tenth and fifteenth postcontrast DISCO images (DISCO-10 and DISCO-15) were selected for further analysis. An experienced radiologist used freely available software (FAE) to perform radiomics extraction. After principal component analysis (PCA), a multilayer perceptron artificial neural network (ANN) to distinguish between malignant and benign lesions was developed and tested using a random allocation approach. ROC analysis was performed to evaluate the diagnostic performance.

Results

173 patients (mean age 43.1 years, range 18-69 years) with 182 lesions (95 benign, 87 malignant) were included. Three types of independent principal components were obtained from the radiomics based on DISCO-10, DISCO-15, and their combination, respectively. In the testing dataset, ANN models showed excellent diagnostic performance with AUC values of 0.915-0.956. Applying the high-sensitivity cutoffs identified in the training dataset demonstrated the potential to reduce the number of unnecessary biopsies by 63.33%-83.33% at the price of one false-negative diagnosis within the testing dataset.

Conclusions

The ultrafast DCE-MRI radiomics-based machine learning model could classify MR BI-RADS category 4 lesions into benign or malignant, highlighting its potential for future application as a new tool for clinical diagnosis.

Deep learning radiomic analysis of DCE-MRI combined with clinical characteristics predicts pathological complete response to neoadjuvant chemotherapy in breast cancer

Objective

The aim of this study was to develop and validate a deep learning-based radiomic (DLR) model combined with clinical characteristics for predicting pathological complete response (pCR) to neoadjuvant chemotherapy (NAC) in breast cancer. For early prediction of pCR, the DLR model was based on pre-treatment and early treatment dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) data.

Materials and methods

This retrospective study included 95 women (mean age, 48.1 years; range, 29-77 years) who underwent DCE-MRI before (pre-treatment) and after two or three cycles of NAC (early treatment) from 2018 to 2021. The patients in this study were randomly divided into a training cohort (n=67) and a validation cohort (n=28) at a ratio of 7:3. Deep learning and handcrafted features were extracted from pre- and early treatment DCE-MRI contoured lesions. These features contribute to the construction of radiomic signature RS1 and RS2 representing information from different periods. Mutual information and least absolute shrinkage and selection operator regression were used for feature selection. A combined model was then developed based on the DCE-MRI features and clinical characteristics. The performance of the models was assessed using the area under the receiver operating characteristic curve (AUC) and compared using the DeLong test.

Results

The overall pCR rate was 25.3% (24/95). One radiomic feature and three deep learning features in RS1, five radiomic features and 11 deep learning features in RS2, and five clinical characteristics remained in the feature selection. The performance of the DLR model combining pre- and early treatment information (AUC=0.900) was better than that of RS1 (AUC=0.644, P=0.068) and slightly higher that of RS2 (AUC=0.888, P=0.604) in the validation cohort. The combined model including pre- and early treatment information and clinical characteristics showed the best ability with an AUC of 0.925 in the validation cohort.

Conclusion

The combined model integrating pre-treatment, early treatment DCE-MRI data, and clinical characteristics showed good performance in predicting pCR to NAC in patients with breast cancer. Early treatment DCE-MRI and clinical characteristics may play an important role in evaluating the outcomes of NAC by predicting pCR.

Robustness of radiomics to variations in segmentation methods in multimodal brain MRI

Radiomics in neuroimaging uses fully automatic segmentation to delineate the anatomical areas for which radiomic features are computed. However, differences among these segmentation methods affect radiomic features to an unknown extent. A scan-rescan dataset (n = 46) of T1-weighted and diffusion tensor images was used. Subjects were split into a sleep-deprivation and a control group. Scans were segmented using four segmentation methods from which radiomic features were computed. First, we measured segmentation agreement using the Dice-coefficient. Second, robustness and reproducibility of radiomic features were measured using the intraclass correlation coefficient (ICC). Last, difference in predictive power was assessed using the Friedman-test on performance in a radiomics-based sleep deprivation classification application. Segmentation agreement was generally high (interquartile range = 0.77–0.90) and median feature robustness to segmentation method variation was higher (ICC > 0.7) than scan-rescan reproducibility (ICC 0.3–0.8). However, classification performance differed significantly among segmentation methods (p < 0.001) ranging from 77 to 84%. Accuracy was higher for more recent deep learning-based segmentation methods. Despite high agreement among segmentation methods, subtle differences significantly affected radiomic features and their predictive power. Consequently, the effect of differences in segmentation methods should be taken into account when designing and evaluating radiomics-based research methods.

Radiomics for pseudoprogression prediction in high grade gliomas: added value of MR contrast agent

Objective

Material & methods

Results

Conclusion

•

Radiomics allows for prediction of pseudoprogression in high-grade gliomas.

•

Use of contrast media boosts the performance of the Radiomics prediction model.

The stability of oncologic MRI radiomic features and the potential role of deep learning: a review

The use of MRI radiomic models for the diagnosis, prognosis and treatment response prediction of tumors has been increasingly reported in literature. However, its widespread adoption in clinics is hampered by issues related to features stability. In the MRI radiomic workflow, the main factors that affect radiomic features computation can be found in the image acquisition and reconstruction phase, in the image pre-processing steps, and in the segmentation of the region of interest on which radiomic indices are extracted. Deep Neural Networks (DNNs), having shown their potentiality in the medical image processing and analysis field, can be seen as an attractive strategy to partially overcome the issues related to radiomic stability and mitigate their impact. In fact, DNN approaches can be prospectively integrated in the MRI radiomic workflow to improve image quality, obtain accurate and reproducible segmentations and generate standardized images. In this review, DNN methods that can be included in the image processing steps of the radiomic workflow are described and discussed, in the light of a detailed analysis of the literature in the context of MRI radiomic reliability.

Combined Diagnostic Accuracy of Diffusion and Perfusion MR Imaging to Differentiate Radiation-Induced Necrosis from Recurrence in Glioblastoma

We aimed to use quantitative values derived from perfusion and diffusion-weighted MR imaging (PWI and DWI) to differentiate radiation-induced necrosis (RIN) from tumor recurrence in Glioblastoma (GBM) and investigate the best parameters for improved diagnostic accuracy and clinical decision-making. Methods: A retrospective analysis of follow-up MRI with new enhancing observations was performed in histopathologically confirmed subjects of post-treated GBM, who underwent re-surgical exploration. Quantitative estimation of rCBV (relative cerebral blood volume) from PWI and three methods of apparent diffusion coefficient (ADC) estimation were performed, namely ADC R1 (whole cross-sectional area of tumor), ADC R2 (only solid enhancing lesion), and ADC R3 (central necrosis). ROC curve and logistic regression analysis was completed. A confusion matrix table created using Excel provided the best combination parameters to ameliorate false-positive and false-negative results. Results: Forty-four subjects with a mean age of 46 years (range, 19−70 years) underwent re-surgical exploration with RIN in 28 (67%) and recurrent tumor in 16 (33%) on histopathology. rCBV threshold of >3.4 had the best diagnostic accuracy (AUC = 0.93, 81% sensitivity and 89% specificity). A multiple logistic regression model showed significant contributions from rCBV (p < 0.001) and ADC R3 (p = 0.001). After analysis of confusion matrix ADC R3 > 2032 × 10−6 mm2 achieved 100% specificity with gain in sensitivity (94% vs. 56%). Conclusions: A combination of parameters had better diagnostic performance, and a stepwise combination of rCBV and ADC R3 obviated unnecessary biopsies in 10% (3/28), leading to improved clinical decision-making.

Treatment Response and Prognosis Evaluation in High-Grade Glioma: An Imaging Review Based on MRI

In recent years, the development of advanced magnetic resonance imaging (MRI) technology and machine learning (ML) have created new tools for evaluating treatment response and prognosis of patients with high-grade gliomas (HGG); however, patient prognosis has not improved significantly. This is mainly due to the heterogeneity between and within HGG tumors, resulting in standard treatment methods not benefitting all patients. Moreover, the survival of patients with HGG is not only related to tumor cells, but also to noncancer cells in the tumor microenvironment (TME). Therefore, during preoperative diagnosis and follow-up treatment of patients with HGG, noninvasive imaging markers are needed to characterize intratumoral heterogeneity, and then to evaluate treatment response and predict prognosis, timeously adjust treatment strategies, and achieve individualized diagnosis and treatment. In this review, we summarize the research progress of conventional MRI, advanced MRI technology, and ML in evaluation of treatment response and prognosis of patients with HGG. We further discuss the significance of the TME in the prognosis of HGG patients, associate imaging features with the TME, indirectly reflecting the heterogeneity within the tumor, and shifting treatment strategies from tumor cells alone to systemic therapy of the TME, which may be a major development direction in the future. LEVEL OF EVIDENCE: 5 TECHNICAL EFFICACY STAGE: 4.

Imaging Biomarkers of Glioblastoma Treatment Response: A Systematic Review and Meta-Analysis of Recent Machine Learning Studies

OBJECTIVE

Monitoring biomarkers using machine learning (ML) may determine glioblastoma treatment response. We systematically reviewed quality and performance accuracy of recently published studies.

METHODS

Following Preferred Reporting Items for Systematic Reviews and Meta-Analysis: Diagnostic Test Accuracy, we extracted articles from MEDLINE, EMBASE and Cochrane Register between 09/2018-01/2021. Included study participants were adults with glioblastoma having undergone standard treatment (maximal resection, radiotherapy with concomitant and adjuvant temozolomide), and follow-up imaging to determine treatment response status (specifically, distinguishing progression/recurrence from progression/recurrence mimics, the target condition). Using Quality Assessment of Diagnostic Accuracy Studies Two/Checklist for Artificial Intelligence in Medical Imaging, we assessed bias risk and applicability concerns. We determined test set performance accuracy (sensitivity, specificity, precision, F1-score, balanced accuracy). We used a bivariate random-effect model to determine pooled sensitivity, specificity, area-under the receiver operator characteristic curve (ROC-AUC). Pooled measures of balanced accuracy, positive/negative likelihood ratios (PLR/NLR) and diagnostic odds ratio (DOR) were calculated. PROSPERO registered (CRD42021261965).

RESULTS

Eighteen studies were included (1335/384 patients for training/testing respectively). Small patient numbers, high bias risk, applicability concerns (particularly confounding in reference standard and patient selection) and low level of evidence, allow limited conclusions from studies. Ten studies (10/18, 56%) included in meta-analysis gave 0.769 (0.649-0.858) sensitivity [pooled (95% CI)]; 0.648 (0.749-0.532) specificity; 0.706 (0.623-0.779) balanced accuracy; 2.220 (1.560-3.140) PLR; 0.366 (0.213-0.572) NLR; 6.670 (2.800-13.500) DOR; 0.765 ROC-AUC.

CONCLUSION

ML models using MRI features to distinguish between progression and mimics appear to demonstrate good diagnostic performance. However, study quality and design require improvement.

OUP accepted manuscript

The recent, rapid advances in immuno-oncology have revolutionized cancer treatment and spurred further research into tumor biology. Yet, cancer patients respond variably to immunotherapy despite mounting evidence to support its efficacy. Current methods for predicting immunotherapy response are unreliable, as these tests cannot fully account for tumor heterogeneity and microenvironment. An improved method for predicting response to immunotherapy is needed. Recent studies have proposed radiomics—the process of converting medical images into quantitative data (features) that can be processed using machine learning algorithms to identify complex patterns and trends—for predicting response to immunotherapy. Because patients undergo numerous imaging procedures throughout the course of the disease, there exists a wealth of radiological imaging data available for training radiomics models. And because radiomic features reflect cancer biology, such as tumor heterogeneity and microenvironment, these models have enormous potential to predict immunotherapy response more accurately than current methods. Models trained on preexisting biomarkers and/or clinical outcomes have demonstrated potential to improve patient stratification and treatment outcomes. In this review, we discuss current applications of radiomics in oncology, followed by a discussion on recent studies that use radiomics to predict immunotherapy response and toxicity.

Radiomics feature reliability assessed by intraclass correlation coefficient: a systematic review

Radiomics research is rapidly growing in recent years, but more concerns on radiomics reliability are also raised. This review attempts to update and overview the current status of radiomics reliability research in the ever expanding medical literature from the perspective of a single reliability metric of intraclass correlation coefficient (ICC). To conduct this systematic review, Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines were followed. After literature search and selection, a total of 481 radiomics studies using CT, PET, or MRI, covering a wide range of subject and disease types, were included for review. In these highly heterogeneous studies, feature reliability to image segmentation was much more investigated than reliability to other factors, such as image acquisition, reconstruction, post-processing, and feature quantification. The reported ICCs also suggested high radiomics feature reliability to image segmentation. Image acquisition was found to introduce much more feature variability than image segmentation, in particular for MRI, based on the reported ICC values. Image post-processing and feature quantification yielded different levels of radiomics reliability and might be used to mitigate image acquisition-induced variability. Some common flaws and pitfalls in ICC use were identified, and suggestions on better ICC use were given. Due to the extremely high study heterogeneities and possible risks of bias, the degree of radiomics feature reliability that has been achieved could not yet be safely synthesized or derived in this review. More future researches on radiomics reliability are warranted.

Comparable Performance of Deep Learning-Based to Manual-Based Tumor Segmentation in KRAS/NRAS/BRAF Mutation Prediction With MR-Based Radiomics in Rectal Cancer

Radiomic features extracted from segmented tumor regions have shown great power in gene mutation prediction, while deep learning–based (DL-based) segmentation helps to address the inherent limitations of manual segmentation. We therefore investigated whether deep learning–based segmentation is feasible in predicting KRAS/NRAS/BRAF mutations of rectal cancer using MR-based radiomics. In this study, we proposed DL-based segmentation models with 3D V-net architecture. One hundred and eight patients’ images (T2WI and DWI) were collected for training, and another 94 patients’ images were collected for validation. We evaluated the DL-based segmentation manner and compared it with the manual-based segmentation manner through comparing the gene prediction performance of six radiomics-based models on the test set. The performance of the DL-based segmentation was evaluated by Dice coefficients, which are 0.878 ± 0.214 and 0.955 ± 0.055 for T2WI and DWI, respectively. The performance of the radiomics-based model in gene prediction based on DL-segmented VOI was evaluated by AUCs (0.714 for T2WI, 0.816 for DWI, and 0.887 for T2WI+DWI), which were comparable to that of corresponding manual-based VOI (0.637 for T2WI, P=0.188; 0.872 for DWI, P=0.181; and 0.906 for T2WI+DWI, P=0.676). The results showed that 3D V-Net architecture could conduct reliable rectal cancer segmentation on T2WI and DWI images. All-relevant radiomics-based models presented similar performances in KRAS/NRAS/BRAF prediction between the two segmentation manners.