A computer-aided tool for automatic volume estimation of hematoma using non-contrast brain CT scans.

The computation of hematoma volume is the key parameter for treatment planning of Intracerebral hemorrhage (ICH). Non-contrast computed tomography (NCCT) imaging is routinely used for the diagnosis of ICH. Hence, the development of computer-aided tools for three-dimensional (3D) computed tomography (CT) image analysis is essential to estimate the gross volume of hematoma. We propose a methodology for automatic estimation of the hematoma volume from 3D CT volumes. Our approach integrates two different methods, multiple abstract splitting (MAS) and seeded region growing (SRG) to develop a unified hematoma detection pipeline from pre-processed CT volumes. The proposed methodology was tested on 80 cases. The volume was estimated from the delineated hematoma region, validated against the ground-truth volumes, and compared with those obtained from the conventional ABC/2 approach. We also compared our results with the U-Net model (supervised technique) to show the applicability of the proposed method. The volume calculated from manually segmented hematoma was considered the ground truth. TheR2correlation coefficient between the volume obtained from the proposed algorithm and the ground truth is 0.86, which is equivalent to theR2value resulting from the comparison between the volume calculated by ABC/2 and the ground truth. The experimental results of the proposed unsupervised approach are comparable to the deep neural architecture (U-Net models). The average computation time was 132.76 ± 14 seconds. The proposed methodology provides a fast and automatic estimation of hematoma volume, which is similar to the baseline user-guided ABC/2 approach. Implementation of our method does not demand a high-end computational setup. Thus, recommended in clinical practice for computer-assistive volume estimation of hematoma from 3D CT volumes and can be implemented in a simple computer system.

Lung Nodule Detection from Feature Engineering to Deep Learning in Thoracic CT Images: a Comprehensive Review.

This paper presents a systematic review of the literature focused on the lung nodule detection in chest computed tomography (CT) images. Manual detection of lung nodules by the radiologist is a sequential and time-consuming process. The detection is subjective and depends on the radiologist's experiences. Owing to the variation in shapes and appearances of a lung nodule, it is very difficult to identify the proper location of the nodule from a huge number of slices generated by the CT scanner. Small nodules (< 10 mm in diameter) may be missed by this manual detection process. Therefore, computer-aided diagnosis (CAD) system acts as a "second opinion" for the radiologists, by making final decision quickly with higher accuracy and greater confidence. The goal of this survey work is to present the current state of the artworks and their progress towards lung nodule detection to the researchers and readers in this domain. This review paper has covered the published works from 2009 to April 2018. Different nodule detection approaches are described elaborately in this work. Recently, it is observed that deep learning (DL)-based approaches are applied extensively for nodule detection and characterization. Therefore, emphasis has been given to convolutional neural network (CNN)-based DL approaches by describing different CNN-based networks.

An association study of severity of intellectual disability with peripheral biomarkers of disabled children in a rehabilitation home, Kolkata, India.

The current investigation has identified the biomarkers associated with severity of disability and correlation among plethora of systemic, cellular and molecular parameters of intellectual disability (ID) in a rehabilitation home. The background of study lies with the recent clinical evidences which identified complications in ID. Various indicators from blood and peripheral system serve as potential surrogates for disability related changes in brain functions. ID subjects (Male, age 10 ± 5 yrs, N = 45) were classified as mild, moderate and severe according to the severity of disability using standard psychometric analysis. Clinical parameters including stress biomarkers, neurotransmitters, RBC morphology, expressions of inflammatory proteins and neurotrophic factor were estimated from PBMC, RBC and serum. The lipid peroxidation of PBMC and RBC membranes, levels of serum glutamate, serotonin, homocysteine, ROS, lactate and LDH-A expression increased significantly with severity of ID whereas changes in RBC membrane ß-actin, serum BDNF, TNF-α and IL-6 was found non-significant. Structural abnormalities of RBC were more in severely disabled children compared to mildly affected ones. The oxidative stress remained a crucial factor with severity of disability. This is confirmed not only by RBC alterations but also with other cellular dysregulations. The present article extends unique insights of how severity of disability is correlated with various clinical, cellular and molecular markers of blood. This unique study primarily focuses on the strong predictors of severity of disability and their associations via brain-blood axis.

Automated in vivo delineation of lumen wall using intravascular ultrasound imaging.

This paper presents a novel methodology for automated detection and extraction of the lumen wall from Intravascular Ultrasound (IVUS) frames. IVUS is an in-vivo pull back imaging technique and provides a sequential frame of images for diagnosis of atherosclerotic heart disease. The detection and segmentation of lumen wall is necessary for predicting the arterial wall blockage. Lumen wall is recognized and segmented with the help of seed refinement and random walks algorithms, in tunica and lumen area. The proposed methodology was tested on 147 frames of 13 patients. Proposed method achieves significant performances for automated lumen wall detection and extraction as compared with existing literature.