Toward an interpretable Alzheimer's disease diagnostic model with regional abnormality representation via deep learning.

In this paper, we propose a novel method for magnetic resonance imaging based Alzheimer's disease (AD) or mild cognitive impairment (MCI) diagnosis that systematically integrates voxel-based, region-based, and patch-based approaches into a unified framework. Specifically, we parcellate the brain into predefined regions based on anatomical knowledge (i.e., templates) and derive complex nonlinear relationships among voxels, whose intensities denote volumetric measurements, within each region. Unlike existing methods that use cubical or rectangular shapes, we consider the anatomical shapes of regions as atypical patches. Using complex nonlinear relationships among voxels in each region learned by deep neural networks, we extract a "regional abnormality representation." We then make a final clinical decision by integrating the regional abnormality representations over the entire brain. It is noteworthy that the regional abnormality representations allow us to interpret and understand the symptomatic observations of a subject with AD or MCI by mapping and visualizing these observations in the brain space. On the baseline MRI dataset from the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort, our method achieves state-of-the-art performance for four binary classification tasks and one three-class classification task. Additionally, we conducted exhaustive experiments and analysis to validate the efficacy and potential of our method.

Multi-scale gradual integration CNN for false positive reduction in pulmonary nodule detection.

Lung cancer is a global and dangerous disease, and its early detection is crucial for reducing the risks of mortality. In this regard, it has been of great interest in developing a computer-aided system for pulmonary nodules detection as early as possible on thoracic CT scans. In general, a nodule detection system involves two steps: (i) candidate nodule detection at a high sensitivity, which captures many false positives and (ii) false positive reduction from candidates. However, due to the high variation of nodule morphological characteristics and the possibility of mistaking them for neighboring organs, candidate nodule detection remains a challenge. In this study, we propose a novel Multi-scale Gradual Integration Convolutional Neural Network (MGI-CNN), designed with three main strategies: (1) to use multi-scale inputs with different levels of contextual information, (2) to use abstract information inherent in different input scales with gradual integration, and (3) to learn multi-stream feature integration in an end-to-end manner. To verify the efficacy of the proposed network, we conducted exhaustive experiments on the LUNA16 challenge datasets by comparing the performance of the proposed method with state-of-the-art methods in the literature. On two candidate subsets of the LUNA16 dataset, i.e., V1 and V2, our method achieved an average CPM of 0.908 (V1) and 0.942 (V2), outperforming comparable methods by a large margin. Our MGI-CNN is implemented in Python using TensorFlow and the source code is available from https://github.com/ku-milab/MGICNN.

The Association between Helicobacter pylori Infection and Body Weight among Children.

PURPOSE: We performed to reveal the association between the Helicobacter pylori infection and body weight among children. METHODS: Out retrospective study included patients who underwent the H. pylori immunoglobulin G testing at Konyang University Hospital between March 2011 and June 2014. These patients were classified as seropositive (28 boys, 27 girls; mean age: 9.89±3.28 years) or seronegative (55 boys, 54 girls; mean age: 9.84±3.02 years). Next, we compared various characteristics between the seropositive and negative groups, as well as between obese children (body weight ≥90th percentile) and non-obese children (body weight <90th percentile). Furthermore, we compared the change in body weight after 2 months of treatment with amoxicillin, clarithromycin and omeprazole among the 55 seropositive children (14 treated children and 41 non-treated children). RESULTS: There were no differences in the weights and laboratory data for the 55 seropositive children and 109 seronegative children (weight; 40.96±18.11 kg vs. 36.85±13.72 kg, respectively; p=0.14). And, there was no difference in the prevalence of H. pylori infection among the 29 obese and 135 non-obese children (p=0.581). However, after 2 months of eradication, the 14 treated patients exhibited a significant weight gain (+0.91±0.52 kg), compared to the 41 non-treated patients (-0.29±1.16 kg, p=0.025). CONCLUSION: Our findings present that obesity was not associated with the H. pylori infection, although H. pylori eradication led to significant increase in body weight.

The Relationships between Respiratory Virus Infection and Aminotransferase in Children.

PURPOSE: We sought to examine the relationship between the clinical manifestations of nonspecific reactive hepatitis and respiratory virus infection in pediatric patients. METHODS: Patients admitted to the pediatric unit of Konyang University Hospital for lower respiratory tract disease between January 1, 2014 and December 31, 2014 and who underwent reverse transcriptase polymerase chain reaction tests were examined. The patients were divided into those with increased levels of alanine aminotransferase (ALT) or aspartate aminotransferase (AST) and those with normal ALT or AST levels. Further, patients with increased ALT and AST levels were individually compared with patients in the normal group, and the blood test results were compared according to the type of respiratory virus. RESULTS: Patients with increased ALT or AST levels had one more day of hospital stay, on average, compared with patients in the normal group (5.3±3.1 days vs. 4.4±3.0 days, p=0.019). Patients in the increased ALT level group were younger and had a longer mean hospital stay, compared with patients in the normal group (p=0.022 and 0.003, respectively). The incidences of increased ALT or AST were the highest in adenovirus infections (6/24, 25.0%), followed by enterovirus (2/11, 18.2%) and respiratory syncytial virus A (21/131, 16.0%) infections. CONCLUSION: Nonspecific reactive hepatitis is more common among patients with adenovirus, enterovirus and respiratory syncytial virus infection, as well as among those infected at a younger age. Compared with AST levels, ALT levels are better indicators of the severity of nonspecific reactive hepatitis.