This research comprises experiments with a deep learning framework for fully automating the skull stripping from brain magnetic resonance (MR) images. Conventional techniques for segmentation have progressed to the extent of Convolutional Neural Networks (CNN). We proposed and experimented with a contemporary variant of the deep learning framework based on mask region convolutional neural network (Mask–RCNN) for all anatomical orientations of brain MR images. We trained the system from scratch to build a model for classification, detection, and segmentation. It is validated by images taken from three different datasets: BrainWeb; NAMIC, and a local hospital. We opted for purposive sampling to select 2000 images of T1 modality from data volumes followed by a multi-stage random sampling technique to segregate the dataset into three batches for training (75%), validation (15%), and testing (10%) respectively. We utilized a robust backbone architecture, namely ResNet–101 and Functional Pyramid Network (FPN), to achieve optimal performance with higher accuracy. We subjected the same data to two traditional methods, namely Brain Extraction Tools (BET) and Brain Surface Extraction (BSE), to compare their performance results. Our proposed method had higher mean average precision (mAP) = 93% and content validity index (CVI) = 0.95%, which were better than comparable methods. We contributed by training Mask–RCNN from scratch for generating reusable learning weights known as transfer learning. We contributed to methodological novelty by applying a pragmatic research lens, and used a mixed method triangulation technique to validate results on all anatomical modalities of brain MR images. Our proposed method improved the accuracy and precision of skull stripping by fully automating it and reducing its processing time and operational cost and reliance on technicians. This research study has also provided grounds for extending the work to the scale of explainable artificial intelligence (XAI).

Purpose: To improve accuracy of apparent diffusion coefficient (ADC) measurement across different bone‐marrow (BM) sites for myelofibrosis (MF) patients. Methods: Vendor‐provided ADC gradient nonlinearity correction (GNC) was implemented for 41 MF study subjects on a 3T clinical scanner. The degree of bias correction was assessed on b‐maps across the BM regions‐of‐interest defined for femoral trochanter, posterior ilium and spine vertebrae of all subjects. Bias variability from subject repositioning was evaluated for five subjects with longitudinal scans using Bland–Altman analysis. BM ADC with and without GNC were compared in 22 subjects with MF grade > 0 (iliac fat fraction < 40%). The GNC effect on ADC heterogeneity trends across bone sites was assessed using paired t‐test and correlation to MF‐grade. Results: The observed bias was substantial across BM sites ranging from −10% for edge vertebrae to +8.4% for trochanter across all subjects with a 13.7% intra‐subject median range. The effect of irreproducible positioning in longitudinal scans was ±4.2% bias (95% limits‐of‐agreement) with range of ±8.5% (highest for L1 and L5). By removing bias, GNC improved ADC accuracy and longitudinal reproducibility and emphasized biological heterogeneity for femoral trochanter versus ilium and edge vertebrae (T11‐L1, L5, and S1). GNC revealed significant ADC differences between trochanters and edge vertebrae ( p = 0.007), while edge vertebrae and ilium ADC became more uniform ( p = 0.07), with heterogeneity and values correlated to MF‐grade. Conclusion: On‐scanner correction of gradient nonlinearity bias in bone marrow ADC reduces technical nonuniformity and variability and allows accurate and reproducible characterization of heterogeneous disease for myelofibrosis patients.