Malignancy rate of lesions presenting as architectural distortion on DBT related to accompanied features, ultrasound findings, and BI-RADS density

European Journal of Radiology · 2025-08-12

MRI Evaluation of Cerebral Microbleeds, Silent Infarct, Iron Deposit with QSM in Sickle Cell Disease and Thalassemia Compared to Healthy Controls

Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition · 2025-09-16

Motivation: Sickle cell disease (SCD) and Thalassemia are inherited blood disorders, and patients are likely to have small vessel disease, silent infarct, and cerebral microbleeds. Goal(s): Multi-parametric MRI was used to evaluate clinically significant findings, and QSM was used to measure iron deposits in various brain regions. Approach: A prospective study was performed by enrolling 30 subjects. The ROIs in the basal ganglia regions were segmented and mapped to QSM to measure susceptivity. Results: Two sickle cell patients had cerebral microbleeds, and two had silent infarcts. The QSM susceptibility was distributed in a wide range, especially in patient groups. Impact: Sickle cell and Thalassemia patients may need blood transfusions, which leads to iron overload. Brain iron may also increase due to microvasculature damage. QSM may be performed to evaluate iron deposit levels and microbleeds to help improve their precision care.

CT-Based deep learning platform combined with clinical parameters for predicting different discharge outcome in spontaneous intracerebral hemorrhage

Neurological Sciences · 2025-09-16

Prediction of High and Low Expression of Tumor-Infiltrating Lymphocytes in Breast Cancer Using MRI Features Combined with Molecular Subtypes

Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition · 2025-09-16

Motivation: Tumor-infiltration lymphocytes (TILs) is a key prognostic factor for breast cancer (BC). Goal(s): To differentiate high vs. low TILs using MRI features, histology, and molecular subtypes. Approach: MRI features were reviewed by radiologists using BI-RADS lexicon. The combined models were built using nomogram and machine learning (ML) algorithms. Results: The classification model built based on MRI features showed AUC of 0.81 and 0.75 in training and testing data, respectively. When combining MRI and clinical parameters, the nomogram showed AUC of 0.82 and 0.79, and the SVM ML model had the best performance, showing an AUC of 0.86 and 0.80. Impact: MRI features could predict high vs. low TILs expression. The three molecular subtypes (HR+/HER2-, HER2+, triple-negative) had distinctly different TILs, and more sophisticated models by combining MRI features with clinical and histological information could improve the TILs prediction accuracy.

Multimodal Diagnosis of Breast Lesions Presenting as Architectural Distortion on DBT by Integrating with Ultrasound and MRI Features

Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition · 2025-09-16

Motivation: Diagnosis of architectural distortion (AD) on DBT is challenging, and supplementary features of breast Ultrasound (US) and MRI may help. Goal(s): To compare the diagnostic performance of models built using the radiological features of DBT, US, MRI, and the combined features. Approach: The BI-RADS categories of DBT, US and MRI were reported, and the machine learning (ML) algorithms were applied to build diagnostic models. Results: DBT, US, and MRI showed comparable diagnostic performance based on BI-RADS categories of 5, 4C, 4B, 4A, 3, 2. The accuracy and AUC were improved using the ML models developed with combined multimodal features. Impact: Diagnosis of lesions presenting as architectural distortion on Digital Breast Tomosynthesis (DBT) is difficult by all breast imaging modalities, and the performance can be improved with ML models developed using the combined multimodal features of DBT, US and MRI.

MR T2* Map to Predict Worsening Hypertension Control: A Preliminary Study

Life · 2025-01-09

Blood pressure measurement is important in monitoring hypertension. However, blood pressure does not provide much information about renal condition in treated hypertension. This study aimed to evaluate renal oxygenation in hypertensive patients using T2* mapping. Subgroup analysis explored whether R2* values can guide adjustments in antihypertensive treatment. A total of 140 consecutive subjects were recruited: 87 hypertensive subjects and 53 normotensive subjects. Hypertensive subjects were classified into non-medication (non-med), angiotensin II receptor blocker (ARB), and non-ARB-treated groups. Each group was divided into good and poor control subgroups based on blood pressure at enrollment. T2* mapping was utilized to assess renal cortical and medullary R2* values. After a 2-year follow-up, subjects were categorized into stable and unstable based on the need for treatment modifications. The unstable subgroup had higher medullary R2* values than the stable subgroup in all followed patients (p < 0.05). Additionally, the unstable merged non-med with ARB subgroup had higher medullary R2* values overall (p < 0.05) and within the good control subgroup (p < 0.05). Patients with stable hypertension, especially those with good control managed through lifestyle modifications or ARBs, exhibited lower renal medullary R2* values, suggesting higher renal oxygenation.

Classification of HER2-zero and HER2-low and Prediction of Progression in Triple Negative Breast Cancer Using MRI Features

Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition · 2025-09-16

Motivation: Triple-negative breast cancer (TNBC) is more aggressive than other molecular subtypes, and finding new biomarkers for targeted therapy and stratification of progression risk would be helpful. Goal(s): To evaluate MRI features between HER2-zero and HER2-low, and between progressed vs. non-progressed TNBC patients. Approach: MRI features were reviewed by radiologists based on BI-RADS lexicon. Results: 85/126 (67%) were identified as HER2-zero and 41 (33%) as HER2-low. No MRI features showed a significant difference between HER2-zero vs HER2-low. Fourteen of 126 (11%) patients showed progression (local, regional progression or distant metastasis), and they have more aggressive finding features and higher stages. Impact: TNBC had negative ER, PR, and HER2, which was defined as immunohistochemical staining 0, 1+, or 2+ with negative FISH, but currently, HER2-low may receive anti-HER2 drug conjugates. Finding imaging biomarkers may help in precision treatment and understanding progression risk.

MRI Features Associated with High and Low Expression of Tumor-Infiltrating Lymphocytes: Stratified Analysis According to Molecular Subtypes

Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition · 2024-11-26

Motivation: Tumor-infiltration lymphocytes (TILs) express variably in different molecular subtypes. Goal(s): To compare the rate of high vs. low TILs and MRI features in three subtypes: Hormonal-Receptor positive, HER2 negative (HR+/HER2-), HER2+, and TN, and compare imaging features in each subtype. Approach: The percentage of TILs of 457 breast cancers was assessed. Three radiologists reviewed MRI features. Results: HER2+ cancers were more likely to present as non-mass enhancement (NME). In HR+, high TILs cases were more likely to present peritumoral edema. In TN, high TILs cases were more likely to present regular shapes and circumscribed margins. Impact: TILs expression increases from HR+ to HER2+ to TN. MRI features in different molecular subtypes show substantial variations. Different models should be built for different subtypes when building MR radiomics models to predict TILs.

Preoperative Differentiation of <scp>HER2‐Zero</scp> and <scp>HER2‐Low</scp> from <scp>HER2‐Positive</scp> Invasive Ductal Breast Cancers Using <scp>BI‐RADS MRI</scp> Features and Machine Learning Modeling

Journal of Magnetic Resonance Imaging · 2024-05-10 · 10 citations

BACKGROUND: Accurate determination of human epidermal growth factor receptor 2 (HER2) is important for choosing optimal HER2 targeting treatment strategies. HER2-low is currently considered HER2-negative, but patients may be eligible to receive new anti-HER2 drug conjugates. PURPOSE: To use breast MRI BI-RADS features for classifying three HER2 levels, first to distinguish HER2-zero from HER2-low/positive (Task-1), and then to distinguish HER2-low from HER2-positive (Task-2). STUDY TYPE: Retrospective. POPULATION: 621 invasive ductal cancer, 245 HER2-zero, 191 HER2-low, and 185 HER2-positive. For Task-1, 488 cases for training and 133 for testing. For Task-2, 294 cases for training and 82 for testing. FIELD STRENGTH/SEQUENCE: 3.0 T; 3D T1-weighted DCE, short time inversion recovery T2, and single-shot EPI DWI. ASSESSMENT: Pathological information and BI-RADS features were compared. Random Forest was used to select MRI features, and then four machine learning (ML) algorithms: decision tree (DT), support vector machine (SVM), k-nearest neighbors (k-NN), and artificial neural nets (ANN), were applied to build models. STATISTICAL TESTS: Chi-square test, one-way analysis of variance, and Kruskal-Wallis test were performed. The P values <0.05 were considered statistically significant. For ML models, the generated probability was used to construct the ROC curves. RESULTS: Peritumoral edema, the presence of multiple lesions and non-mass enhancement (NME) showed significant differences. For distinguishing HER2-zero from non-zero (low + positive), multiple lesions, edema, margin, and tumor size were selected, and the k-NN model achieved the highest AUC of 0.86 in the training set and 0.79 in the testing set. For differentiating HER2-low from HER2-positive, multiple lesions, edema, and margin were selected, and the DT model achieved the highest AUC of 0.79 in the training set and 0.69 in the testing set. DATA CONCLUSION: BI-RADS features read by radiologists from preoperative MRI can be analyzed using more sophisticated feature selection and ML algorithms to build models for the classification of HER2 status and identify HER2-low. TECHNICAL EFFICACY: Stage 2.

Preoperative Prediction of Her2-zero, -low and -overexpression Breast Cancers Using Multiparametric MRI and Machine Learning Modeling

Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition · 2024-11-26

Motivation: Her2-low breast cancers could benefit from new anti-HER2 therapies. Goal(s): To construct a preoperative prediction model of HER2 expression levels using multiparametric MRI and machine learning (ML) algorithms. Approach: 621 patients were investigated. Four ML methods were used to build models based on MRI features to predict HER2 expression levels. Results: MRI features of multiple lesions, spiculated margin, peritumoral edema and largest diameter were selected to build the models. ML models performed better for predicting HER2-zero vs. HER2-low/-overexpression than HER2-low vs. HER2-overexpression. The best model was KNN of AUC 0.86, sensitivity of 76%, specificity of 73%, and accuracy of 75%. Impact: MRI features of breast cancer are associated with different HER2 expression levels. MRI-based ML models have the potential to preoperatively predict the HER2 expression status.