Detection of Acute Infarction on Non-Contrast-enhanced CT: Closing the Gap with MRI via Machine Learning

Machine-Learning and Radiomics-Based Preoperative Prediction of Ki-67 Expression in Glioma Using MRI Data

BACKGROUND

Gliomas are the most common primary brain tumours and constitute approximately half of all malignant glioblastomas. Unfortunately, patients diagnosed with malignant glioblastomas typically survive for less than a year. In light of this circumstance, genotyping is an effective means of categorising gliomas. The Ki67 proliferation index, a widely used marker of cellular proliferation in clinical contexts, has demonstrated potential for predicting tumour classification and prognosis. In particular, magnetic resonance imaging (MRI) plays a vital role in the diagnosis of brain tumours. Using MRI to extract glioma-related features and construct a machine learning model offers a viable avenue to classify and predict the level of Ki67 expression.

METHODS

This study retrospectively collected MRI data and postoperative immunohistochemical results from 613 glioma patients from the First Affliated Hospital of Nanjing Medical University. Subsequently, we performed registration and skull stripping on the four MRI modalities: T1-weighted (T1), T2-weighted (T2), T1-weighted with contrast enhancement (T1CE), and Fluid Attenuated Inversion Recovery (FLAIR). Each modality's segmentation yielded three distinct tumour regions. Following segmentation, a comprehensive set of features encompassing texture, first-order, and shape attributes were extracted from these delineated regions. Feature selection was conducted using the least absolute shrinkage and selection operator (LASSO) algorithm with subsequent sorting to identify the most important features. These selected features were further analysed using correlation analysis to finalise the selection for machine learning model development. Eight models: logistic regression (LR), naive bayes, decision tree, gradient boosting tree, and support vector classification (SVM), random forest (RF), XGBoost, and LightGBM were used to objectively classify Ki67 expression.

RESULTS

In total, 613 patients were enroled in the study, and 24,455 radiomic features were extracted from each patient's MRI. These features were eventually reduced to 36 after LASSO screening, RF importance ranking, and correlation analysis. Among all the tested machine learning models, LR and linear SVM exhibited superior performance. LR achieved the highest area under the curve score of 0.912 ± 0.036, while linear SVM obtained the top accuracy with a score of 0.884 ± 0.031.

CONCLUSION

This study introduced a novel approach for classifying Ki67 expression levels using MRI, which has been proven to be highly effective. With the LR model at its core, our method demonstrated its potential in signalling a promising avenue for future research. This innovative approach of predicting Ki67 expression based on MRI features not only enhances our understanding of cell activity but also represents a significant leap forward in brain glioma research. This underscores the potential of integrating machine learning with medical imaging to aid in the diagnosis and prognosis of complex diseases.

Hybrid CNN-Transformer Network With Circular Feature Interaction for Acute Ischemic Stroke Lesion Segmentation on Non-Contrast CT Scans

Lesion segmentation is a fundamental step for the diagnosis of acute ischemic stroke (AIS). Non-contrast CT (NCCT) is still a mainstream imaging modality for AIS lesion measurement. However, AIS lesion segmentation on NCCT is challenging due to low contrast, noise and artifacts. To achieve accurate AIS lesion segmentation on NCCT, this study proposes a hybrid convolutional neural network (CNN) and Transformer network with circular feature interaction and bilateral difference learning. It consists of parallel CNN and Transformer encoders, a circular feature interaction module, and a shared CNN decoder with a bilateral difference learning module. A new Transformer block is particularly designed to solve the weak inductive bias problem of the traditional Transformer. To effectively combine features from CNN and Transformer encoders, we first design a multi-level feature aggregation module to combine multi-scale features in each encoder and then propose a novel feature interaction module containing circular CNN-to-Transformer and Transformer-to-CNN interaction blocks. Besides, a bilateral difference learning module is proposed at the bottom level of the decoder to learn the different information between the ischemic and contralateral sides of the brain. The proposed method is evaluated on three AIS datasets: the public AISD, a private dataset and an external dataset. Experimental results show that the proposed method achieves Dices of 61.39% and 46.74% on the AISD and the private dataset, respectively, outperforming 17 state-of-the-art segmentation methods. Besides, volumetric analysis on segmented lesions and external validation results imply that the proposed method is potential to provide support information for AIS diagnosis.

Toward Generalizability in the Deployment of Artificial Intelligence in Radiology: Role of Computation Stress Testing to Overcome Underspecification

The clinical deployment of artificial intelligence (AI) applications in medical imaging is perhaps the greatest challenge facing radiology in the next decade. One of the main obstacles to the incorporation of automated AI-based decision-making tools in medicine is the failure of models to generalize when deployed across institutions with heterogeneous populations and imaging protocols. The most well-understood pitfall in developing these AI models is overfitting, which has, in part, been overcome by optimizing training protocols. However, overfitting is not the only obstacle to the success and generalizability of AI. Underspecification is also a serious impediment that requires conceptual understanding and correction. It is well known that a single AI pipeline, with prescribed training and testing sets, can produce several models with various levels of generalizability. Underspecification defines the inability of the pipeline to identify whether these models have embedded the structure of the underlying system by using a test set independent of, but distributed identically, to the training set. An underspecified pipeline is unable to assess the degree to which the models will be generalizable. Stress testing is a known tool in AI that can limit underspecification and, importantly, assure broad generalizability of AI models. However, the application of stress tests is new in radiologic applications. This report describes the concept of underspecification from a radiologist perspective, discusses stress testing as a specific strategy to overcome underspecification, and explains how stress tests could be designed in radiology-by modifying medical images or stratifying testing datasets. In the upcoming years, stress tests should become in radiology the standard that crash tests have become in the automotive industry. Keywords: Computer Applications-General, Informatics, Computer-aided Diagnosis © RSNA, 2021.

Image Patch-Based Net Water Uptake and Radiomics Models Predict Malignant Cerebral Edema After Ischemic Stroke

Malignant cerebral edema (MCE) after an ischemic stroke results in a poor outcome or death. Early prediction of MCE helps to identify subjects that could benefit from a surgical decompressive craniectomy. Net water uptake (NWU) in an ischemic lesion is a predictor of MCE; however, CT perfusion and lesion segmentation are required. This paper proposes a new Image Patch-based Net Water Uptake (IP-NWU) procedure that only uses non-enhanced admission CT and does not need lesion segmentation. IP-NWU is calculated by comparing the density of ischemic and contralateral normal patches selected from the middle cerebral artery (MCA) area using standard reference images. We also compared IP-NWU with the Segmented Region-based NWU (SR-NWU) procedure in which segmented ischemic regions from follow-up CT images are overlaid onto admission images. Furthermore, IP-NWU and its combination with imaging features are used to construct predictive models of MCE with a radiomics approach. In total, 116 patients with an MCA infarction (39 with MCE and 77 without MCE) were included in the study. IP-NWU was significantly higher for patients with MCE than those without MCE (p < 0.05). IP-NWU can predict MCE with an AUC of 0.86. There was no significant difference between IP-NWU and SR-NWU, nor between their predictive efficacy for MCE. The inter-reader and interoperation agreement of IP-NWU was exceptional according to the Intraclass Correlation Coefficient (ICC) analysis (inter-reader: ICC = 0.92; interoperation: ICC = 0.95). By combining IP-NWU with imaging features through a random forest classifier, the radiomics model achieved the highest AUC (0.96). In summary, IP-NWU and radiomics models that combine IP-NWU with imaging features can precisely predict MCE using only admission non-enhanced CT images scanned within 24 h from onset.