Diagnostic performance of hippocampal volumetry in Alzheimer’s disease or mild cognitive impairment: a meta-analysis

Park, Ho Young; Suh, Chong Hyun; Heo, Hwon; Shim, Woo Hyun; Kim, Sang Joon

doi:10.1007/s00330-022-08838-9

Cited by 13 publications

(10 citation statements)

References 65 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The majority of current studies ( Liu et al, 2020 ; Nadal et al, 2020 ; Park et al, 2022 ) used deep learning networks in AD classification while we took the classical classifiers in machine learning, which is more interpretable and can find radiomics features associated with AD. Radiomic analysis has been applied to some AD studies.…”

Section: Discussionmentioning

confidence: 99%

“…In the field of medical imaging, deep learning-based medical imaging applications outperform traditional methods in complex tasks ( Duraisamy et al, 2019 ; Singh et al, 2020 ). Of all brain regions associated with AD, the hippocampus is one of the earliest to undergo pathology changes ( Han et al, 2014 ; Dhikav et al, 2017 ; Park et al, 2022 ). Helaly et al (2022) proposed a deep learning Alzheimer’s disease hippocampus segmentation framework (DL-AHS) for hippocampal segmentation to detect and identify AD.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A novel cascade machine learning pipeline for Alzheimer’s disease identification and prediction

Zhou

Piao²,

Liu³

et al. 2023

Front. Aging Neurosci.

View full text Add to dashboard Cite

IntroductionAlzheimer’s disease (AD) is a progressive and irreversible brain degenerative disorder early. Among all diagnostic strategies, hippocampal atrophy is considered a promising diagnostic method. In order to proactively detect patients with early Alzheimer’s disease, we built an Alzheimer’s segmentation and classification (AL-SCF) pipeline based on machine learning.MethodsIn our study, we collected coronal T1 weighted images that include 187 patients with AD and 230 normal controls (NCs). Our pipeline began with the segmentation of the hippocampus by using a modified U2-net. Subsequently, we extracted 851 radiomics features and selected 37 features most relevant to AD by the Hierarchical clustering method and Least Absolute Shrinkage and Selection Operator (LASSO) algorithm. At last, four classifiers were implemented to distinguish AD from NCs, and the performance of the models was evaluated by accuracy, specificity, sensitivity, and area under the curve.ResultsOur proposed pipeline showed excellent discriminative performance of classification with AD vs NC in the training set (AUC=0.97, 95% CI: (0.96-0.98)). The model was also verified in the validation set with Dice=0.93 for segmentation and accuracy=0.95 for classification.DiscussionThe AL-SCF pipeline can automate the process from segmentation to classification, which may assist doctors with AD diagnosis and develop individualized medical plans for AD in clinical practice.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A novel cascade machine learning pipeline for Alzheimer’s disease identification and prediction

Zhou

Piao²,

Liu³

et al. 2023

Front. Aging Neurosci.

View full text Add to dashboard Cite

show abstract

“…ERC volume may be a helpful biomarker of sMRI in the diagnosis of AD dementia. 68 Brain structural changes may occur at preclinical stage which is earlier than MCI. 69 Precuneus cortical thinning compared with normal controls appeared 13 years before the expected onset of AD dementia in ADAD patients, being the first cortical region to show divergence.…”

Section: Structural Mrimentioning

confidence: 99%

“…79 ASL has the problems of individual differences and low signalto-noise ratio, and further development is needed for its wide application. 68…”

Section: 22mentioning

confidence: 99%

Alzheimer's disease: Potential pathogenesis and imaging findings

Zhang

Liu

et al. 2023

VIEW

View full text Add to dashboard Cite

Alzheimer's disease (AD) is a common neurodegenerative disease. The histopathological changes in AD include amyloid β‐protein (Aβ) deposition, tau tangles, neuroinflammation, and neurodegeneration. Some of the pathological changes could be shown in vivo by positron emission tomography (PET) and magnetic resonance imaging (MRI) biomarkers, which play a key role in diagnosing AD. Fluorodeoxyglucose positron emission tomography (FDG‐PET) can reflect and predict dysfunction. Aβ‐PET is sensitive for the diagnosis of early AD but cannot distinguish the severity of AD. Tau‐PET can compensate for the deficiency of Aβ‐PET. Tau tangles are positively correlated with the severity of AD and associated with cognitive impairment. Probes targeting neuroinflammation in AD have been developed, but further study is needed to validate their effectiveness. Conventional MRI performs high tissue contrast that can show structural changes and has been routinely applied in clinical practice, such as in the evaluation of cerebral atrophy. Advanced MRI sequences (such as diffusion tensor imaging, arterial spin labeling, magnetic resonance spectroscopy, blood oxygenation level dependent, and quantitative susceptibility mapping) can provide additional information beyond structure that includes brain microstructure, blood perfusion, metabolite concentration, brain activity, connections and networks between brain regions, iron deposition, etc. Integrated PET and MRI may improve the diagnostic efficiency of AD.

show abstract

“…이후 반자동 혹은 자동화 분할에 의한 해마 용적측정의에 대한 연구가 많이 되었고 33개 논문을 기반으로한 메타분석에서 해마 부피를 이용한 정상인 대비 알츠하이머성 치매 진단능이 민감도 82%, 특이도 87%, AUC 0.91로 보고되었다( 11 ). 하지만 연구용뇌용적 분석 소프트웨어는 표준화 문제, 복잡한 사용자 환경, 긴 프로세싱 시간 등의 문제가 있었고( 12 ), 이런 문제점을 극복한 여러 완전 자동화 뇌용적 분석 임상용 소프트웨어들이 미국 식품의약안전청 및 유럽의 승인을 받고 사용 중에 있다.…”

Section: 뇌용적 분석 소프트웨어의 발전unclassified

Brain MRI-Based Artificial Intelligence Software in Patients with Neurodegenerative Diseases: Current Status

Jeong

Suh

Park

et al. 2022

J Korean Soc Radiol

Self Cite

View full text Add to dashboard Cite

현대사회가 점차 고령화 사회가 됨에 따라 퇴행성 뇌질환의 발병률이 증가하고 있으며, 이러한 퇴행성 뇌질환에 관한 많은 연구들이 이루어지고 있다. 퇴행성 뇌질환의 진단에서 영상분석은 영상표지자로서 중요한 역할을 하고 있다. 영상분석에서 객관적이고 일관성 있는 평가는 퇴행성 뇌질환의 조기 진단 및 정확한 진단에 중요하다. 이에 다양한 퇴행성 뇌질환과 관련한 영상연구에 자기공명영상(이하 MRI)을 이용한 인공지능이 조기 진단과 최적의 치료 방향 계획 및 결정에 도움이 될 가능성을 보여주었다. 특히 MRI 기반의 뇌용적 측정과 분획화 및 특성을 포착하는 인공지능 소프트웨어들이 개발되고 연구되기 시작했다. 본 고찰에서는 우리나라에서 퇴행성 뇌질환과 관련하여 사용되고 있는 인공지능 소프트웨어의 현재 상황과 향후 인공지능 소프트웨어의 퇴행성 뇌질환 연구에의 활용, 그리고 인공지능 소프트웨어의 한계에 대해서 다루고자 한다.

show abstract

Diagnostic performance of hippocampal volumetry in Alzheimer’s disease or mild cognitive impairment: a meta-analysis

Cited by 13 publications

References 65 publications

A novel cascade machine learning pipeline for Alzheimer’s disease identification and prediction

A novel cascade machine learning pipeline for Alzheimer’s disease identification and prediction

Alzheimer's disease: Potential pathogenesis and imaging findings

Brain MRI-Based Artificial Intelligence Software in Patients with Neurodegenerative Diseases: Current Status

Contact Info

Product

Resources

About