Search

Multiple instance learning (MIL) is a supervised learning methodology that aims to allow models to learn instance class labels from bag class labels, where a bag is defined to contain multiple instances. MIL is gaining traction for learning from weak labels but has not been widely applied to 3D medical imaging. MIL is well-suited to clinical CT acquisitions since (1) the highly anisotropic voxels hinder application of traditional 3D networks and (2) patch-based networks have limited ability to learn whole volume labels. In this work, we apply MIL with a deep convolutional neural network to identify whether clinical CT head image volumes possess one or more large hemorrhages (> 20cm$^3$), resulting in a learned 2D model without the need for 2D slice annotations. Individual image volumes are considered separate bags, and the slices in each volume are instances. Such a framework sets the stage for incorporating information obtained in clinical reports to help train a 2D segmentation approach. Within this context, we evaluate the data requirements to enable generalization of MIL by varying the amount of training data. Our results show that a training size of at least 400 patient image volumes was needed to achieve accurate per-slice hemorrhage detection. Over a five-fold cross-validation, the leading model, which made use of the maximum number of training volumes, had an average true positive rate of 98.10%, an average true negative rate of 99.36%, and an average precision of 0.9698. The models have been made available along with source code to enabled continued exploration and adaption of MIL in CT neuroimaging.

Brain imaging analysis on clinically acquired computed tomography (CT) is essential for the diagnosis, risk prediction of progression, and treatment of the structural phenotypes of traumatic brain injury (TBI). However, in real clinical imaging scenarios, entire body CT images (e.g., neck, abdomen, chest, pelvis) are typically captured along with whole brain CT scans. For instance, in a typical sample of clinical TBI imaging cohort, only ~15% of CT scans actually contain whole brain CT images suitable for volumetric brain analyses; the remaining are partial brain or non-brain images. Therefore, a manual image retrieval process is typically required to isolate the whole brain CT scans from the entire cohort. However, the manual image retrieval is time and resource consuming and even more difficult for the larger cohorts. To alleviate the manual efforts, in this paper we propose an automated 3D medical image retrieval pipeline, called deep montage-based image retrieval (dMIR), which performs classification on 2D montage images via a deep convolutional neural network. The novelty of the proposed method for image processing is to characterize the medical image retrieval task based on the montage images. In a cohort of 2000 clinically acquired TBI scans, 794 scans were used as training data, 206 scans were used as validation data, and the remaining 1000 scans were used as testing data. The proposed achieved accuracy=1.0, recall=1.0, precision=1.0, f1=1.0 for validation data, while achieved accuracy=0.988, recall=0.962, precision=0.962, f1=0.962 for testing data. Thus, the proposed dMIR is able to perform accurate CT whole brain image retrieval from large-scale clinical cohorts., Accepted for SPIE: Medical Imaging 2019

Medical imaging plays a key role in guiding treatment of traumatic brain injury (TBI) and for diagnosing intracranial hemorrhage; most commonly rapid computed tomography (CT) imaging is performed. Outcomes for patients with TBI are variable and difficult to predict upon hospital admission. Quantitative outcome scales (e.g., the Marshall classification) have been proposed to grade TBI severity on CT, but such measures have had relatively low value in staging patients by prognosis. Herein, we examine a cohort of 1,003 subjects admitted for TBI and imaged clinically to identify potential prognostic metrics using a “big data” paradigm. For all patients, a brain scan was segmented with multi-atlas labeling, and intensity/volume/texture features were computed in a localized manner. In a 10-fold cross-validation approach, the explanatory value of the image-derived features is assessed for length of hospital stay (days), discharge disposition (five point scale from death to return home), and the Rancho Los Amigos functional outcome score (Rancho Score). Image-derived features increased the predictive R2 to 0.38 (from 0.18) for length of stay, to 0.51 (from 0.4) for discharge disposition, and to 0.31 (from 0.16) for Rancho Score (over models consisting only of non-imaging admission metrics, but including positive/negative radiological CT findings). This study demonstrates that high volume retrospective analysis of clinical imaging data can reveal imaging signatures with prognostic value. These targets are suited for follow-up validation and represent targets for future feature selection efforts. Moreover, the increase in prognostic value would improve staging for intervention assessment and provide more reliable guidance for patients.