Manifold Embedding Induced by Multidimensional Scaling and Its Application to Alzheimer’s Disease and Mild Cognitive Impairment

박현진,ADNI 한국물리학회 2011 THE JOURNAL OF THE KOREAN PHYSICAL SOCIETY Vol.59 No.6

Quantifying shape information related to a disease such as Alzheimer’s disease (AD) is an important brain research. Neuroimaging data are high dimensional and thus cumbersome to analyze. Manifold learning techniques, which find a low dimensional representation for high dimensional data, have been applied to brain MRI. A shape quantification method based on multidimensional scaling (MDS), a well known manifold learning technique, was proposed. The method successfully classified between AD and normal. We extended the MDS based quantification method by 1) applying to distinguish mild cognitive impairment (MCI) from normal, and 2) showing the effectiveness of the induced low dimensional embedding space to predict key clinical variables such as mini mental state exam scores and clinical diagnosis using the standard multiple linear regression. Distinguishing MCI from normal is also important as it is related to early detection of AD. We were able to 1) classify not only AD/normal but also MCI/normal better than the traditional classification based on hippocampus volume, and 2) show good statistical power for predicting key clinical variables. Our test group consisted of 25 normal, 25 AD, and 25 MCI patients.

Comparison of Distance Measures for Manifold Learning: Application to Alzheimer’s Brain Scans

박현진,the ADNI 한국물리학회 2012 THE JOURNAL OF THE KOREAN PHYSICAL SOCIETY Vol.61 No.7

The scale of medical imaging data is growing rapidly and automated computer algorithm is well suited to analyze such data. Shape information can distinguish diseased scan from normal controls but it is difficult to analyze the data due to the high dimensionality of shape information. With manifold learning, shape analysis becomes more tractable in the low dimensional space. Some manifold learning methods including multidimensional scaling (MDS) require a distance measure to quantify pair-wise dissimilarity between scans of interest. In this study, we compared two different distance measures combined with MDS to distinguish Alzheimer's disease (AD) and mild cognitive impairment (MCI) from normal control patients. The first distance measure is based on the displacement field and the second distance measure is based on the mutual information (MI). Shape quantification was applied to 25 normal, 25 AD, and 25 MCI patients' brain scans. Use of the first distance measure resulted in 18% error rate, while the second distance measure resulted in 46% error rate for classifying between AD and normal. Application of MDS leads to a feature space and we compared MDS induced feature space with the feature space induced from hippocampus volume, a traditionally used feature for distinguishing AD/MCI from normal.

이중 트리 복합 웨이브렛 변환과 최소 중복과 최대 연관 특징을 이용한 알츠하이머 병 분류

알람 사루알,권구락,ADNI 한국차세대컴퓨팅학회 2016 한국차세대컴퓨팅학회 논문지 Vol.12 No.3

변량 분석 알고리듬은 알츠하이머 병을 예측하기 위한 툴로 종종 사용된다. 정상인으부터 경도인지장애와 알츠하이머 병의 효과적인 조기 예측 및 진단은 매우 중요하다. 조기 예방 치료는 노인 복지에 있어 점차적으로 위험요소를 줄이 는데 도움이 되며 안정적인 삶이 보장되어 질 것이다. 비침습적인 바이오마커로 MRI가 사용된다. 이는 형태학적 차이 와 뇌 위축 정도를 파악한다. 그리고 Freesurfer 툴을 사용하여 피질 분할 후 MRI 슬라이스를 사용한다. 새로운 접 근법은 듀얼 트리 복한 웨이블릿 계수 추출, 최소 중복성과 최대 관계 특색 부분 집합을 사용하여 정상인과 알츠하이 머 병 예측에 적용한다. 새로운 제안 방법의 성능은 91.02±3.38%, 83.10±3.4%로 높은 특이도와 민감도의 성능이 며 예측 정확도는 87.07±2.71%까지 산출되어진다. 실험결과는 제안 방법은 기존의 방법보다 우수함을 보여준다. Multivariate analysis algorithms have been frequently used tool for predicting Alzheimer disease (AD). An efficient early prediction and diagnosis of AD and Mild Cognitive Impairment (MCI) from Healthy Controls (HC) is always very important. Early preventive care could help to degenerate its risk factors gradually in the elderly and the stable life will be guaranteed. The noninvasive biomarker of Magnetic Resonance (MR) images are used here because morphometric difference and cerebral atrophy could be realized. And we also used MR image slices after subcortical segmentation by using FreeSurfer tool. A novel approach is applied for predicting AD from HC using dual tree complex wavelet coefficients extraction, principal components of min-redundancy and max relevance feature subset selection. The prediction accuracy of the novel proposed method is yielded up to 87.07±2.71% with high specificity, sensitivity about 91.02±3.38%, 83.10±3.4% respectively. The experimental results show that the performance of our proposed method is better than that of conventional ones.

Improved Diagnostic Accuracy of Alzheimer’s Disease by Combining Regional Cortical Thickness and Default Mode Network Functional Connectivity: Validated in the Alzheimer’s Disease Neuroimaging Initiative Set

박지은,박범우,김상준,김호성,최충곤,정승채,오주영,이재홍,노지훈,심우현,Alzheimer’s Disease Neuroimaging Initiative (ADNI) 대한영상의학회 2017 Korean Journal of Radiology Vol.18 No.6

Objective: To identify potential imaging biomarkers of Alzheimer’s disease by combining brain cortical thickness (CThk) and functional connectivity and to validate this model’s diagnostic accuracy in a validation set. Materials and Methods: Data from 98 subjects was retrospectively reviewed, including a study set (n = 63) and a validation set from the Alzheimer’s Disease Neuroimaging Initiative (n = 35). From each subject, data for CThk and functional connectivity of the default mode network was extracted from structural T1-weighted and resting-state functional magnetic resonance imaging. Cortical regions with significant differences between patients and healthy controls in the correlation of CThk and functional connectivity were identified in the study set. The diagnostic accuracy of functional connectivity measures combined with CThk in the identified regions was evaluated against that in the medial temporal lobes using the validation set and application of a support vector machine. Results: Group-wise differences in the correlation of CThk and default mode network functional connectivity were identified in the superior temporal (p < 0.001) and supramarginal gyrus (p = 0.007) of the left cerebral hemisphere. Default mode network functional connectivity combined with the CThk of those two regions were more accurate than that combined with the CThk of both medial temporal lobes (91.7% vs. 75%). Conclusion: Combining functional information with CThk of the superior temporal and supramarginal gyri in the left cerebral hemisphere improves diagnostic accuracy, making it a potential imaging biomarker for Alzheimer’s disease.

Serum Concentrations of Selenium and Copper in Patients Diagnosed with Pancreatic Cancer

Marcin R. Lener,Rodney J. Scott,Anna Wiechowska-Koz!owska,Pablo Serrano-Fernández,Piotr Baszuk,Katarzyna Jaworska-Bieniek,Grzegorz Sukiennicki,Wojciech Marciniak,Magdalena Muszy"ska,Józef K!adny,Tomas 대한암학회 2016 Cancer Research and Treatment Vol.48 No.3

Purpose Understanding of the etiology and pathogenesis of pancreatic cancer (PaCa) is still insufficient. This study evaluated the associations between concentrations of selenium (Se) and copper (Cu) in the serum of PaCa patients. Materials and Methods The study included 100 PaCa patients and 100 control subjects from the same geographical region in Poland. To determine the average concentration of Se, Cu, and ratio Cu:Se in the Polish population, assay for Se and Cu was performed in 480 healthy individuals. Serum levels of Se and Cu were measured using inductively coupled plasma mass spectrometry. Results In the control group, the average Se level was 76 !g/L and Cu 1,098 !g/L. The average Se level among PaCa patients was 60 !g/L and the mean Cu level was 1,432 !g/L. The threshold point at which any decrease in Se concentration was associated with PaCa was 67.45 !g/L. The threshold point of Cu level above which there was an increase in the prevalence of PaCa was 1,214.58 !g/L. In addition, a positive relationship was observed between increasing survival time and Se plasma level. Conclusion This retrospective study suggests that low levels of Se and high levels of Cu might influence development of PaCa and that higher levels of Se are associated with longer survival in patients with PaCa. The results suggest that determining the level of Se and Cu could be incorporated into a risk stratification scheme for the selection and surveillance control examination to complement existing screening and diagnostic procedures.