scholarly journals Deep learning from multiple experts improves identification of amyloid neuropathologies

2021 ◽  
Author(s):  
Daniel R. Wong ◽  
Ziqi Tang ◽  
Nicholas C. Mew ◽  
Sakshi Das ◽  
Justin Athey ◽  
...  
Keyword(s):  
Deep Learning ◽  
Cerebral Amyloid Angiopathy ◽  
Expert Consensus ◽  
Amyloid Angiopathy ◽  
Consensus Model ◽  
Entire Cohort ◽  
Novice To Expert ◽  
Cerebral Amyloid ◽  
Multiple Experts

AbstractPathologists can have complementary assessments and focus areas when identifying and labeling neuropathologies. A standardized approach would ideally draw on the expertise of the entire cohort. We present a deep learning (DL) framework that consistently labels cored, diffuse, and cerebral amyloid angiopathy (CAA) neuropathologies using expert consensus. We collected 100,495 annotations, comprising 20,099 candidate neuropathologies from three institutions, independently annotated by five experts. We compared DL methods that learned the annotation behaviors of individual experts (AUPRC=0.67±0.06 cored; 0.48±0.06 CAA) versus those that reproduced expert consensus, yielding 8.9-13% improvements (AUPRC=0.73±0.03 cored; 0.54±0.06 CAA). Saliency mapping on neuropathologies illustrated how human expertise may progress from novice to expert. In blind prospective tests of 52,555 subsequently expert-annotated images, the models accurately labeled pathologies similar to their human counterparts (consensus model AUPRC=0.73 cored; 0.68 CAA).

scholarly journals Deep learning assisted quantitative assessment of histopathological markers of Alzheimer’s disease and cerebral amyloid angiopathy

2021 ◽  
Vol 9 (1) ◽  
Author(s):  
Valentina Perosa ◽  
Ashley A. Scherlek ◽  
Mariel G. Kozberg ◽  
Lindsey Smith ◽  
Thomas Westerling-Bui ◽  
...  
Keyword(s):  
Alzheimer’S Disease ◽  
Alzheimer's Disease ◽  
Deep Learning ◽  
Cerebral Amyloid Angiopathy ◽  
Calcium Deposition ◽  
Amyloid Angiopathy ◽  
Research Centers ◽  
Independent Expert ◽  
Cerebral Amyloid ◽  
Statistical Approaches

AbstractTraditionally, analysis of neuropathological markers in neurodegenerative diseases has relied on visual assessments of stained sections. Resulting semiquantitative scores often vary between individual raters and research centers, limiting statistical approaches. To overcome these issues, we have developed six deep learning-based models, that identify some of the most characteristic markers of Alzheimer’s disease (AD) and cerebral amyloid angiopathy (CAA). The deep learning-based models are trained to differentially detect parenchymal amyloid β (Aβ)-plaques, vascular Aβ-deposition, iron and calcium deposition, reactive astrocytes, microglia, as well as fibrin extravasation. The models were trained on digitized histopathological slides from brains of patients with AD and CAA, using a workflow that allows neuropathology experts to train convolutional neural networks (CNNs) on a cloud-based graphical interface. Validation of all models indicated a very good to excellent performance compared to three independent expert human raters. Furthermore, the Aβ and iron models were consistent with previously acquired semiquantitative scores in the same dataset and allowed the use of more complex statistical approaches. For example, linear mixed effects models could be used to confirm the previously described relationship between leptomeningeal CAA severity and cortical iron accumulation. A similar approach enabled us to explore the association between neuroinflammation and disparate Aβ pathologies. The presented workflow is easy for researchers with pathological expertise to implement and is customizable for additional histopathological markers. The implementation of deep learning-assisted analyses of histopathological slides is likely to promote standardization of the assessment of neuropathological markers across research centers, which will allow specific pathophysiological questions in neurodegenerative disease to be addressed in a harmonized way and on a larger scale.

EVALUATION OF THE UNITED KINGDOM NATIONAL EXTERNAL QUALITY ASSESSMENT SERVICE 2008 (UK NEQAS 2008) GUIDELINES FOR THE DIAGNOSIS OF SUBARACHNOID HEMORRHAGE USING SPECTROPHOTOMETRIC XANTHOCHROMIA: A RETROSPECTIVE STUDY

2009 ◽  
Vol 32 (6S) ◽  
pp. 5
Author(s):  
A Gangloff ◽  
L Nadeau
Keyword(s):  
Subarachnoid Hemorrhage ◽  
Cerebral Amyloid Angiopathy ◽  
Viral Encephalitis ◽  
Objective Evaluation ◽  
Final Diagnosis ◽  
Amyloid Angiopathy ◽  
The United Kingdom ◽  
Cerebral Amyloid ◽  
The One ◽  
The Uk

Objective: Evaluation of the UK NEQAS 2008 guidelines for the interpretation of spectrophotometric xanthochromia. Method: A search of the laboratory database for all the xanthochromia test results between May 1st 2008 and May 1st 2009 was performed. Medical charts were reviewed for patients of Hôpital de l’Enfant-Jésus (HEJ) that had at least one detectable pigment (bilirubin, oxyhemoglobin, or methemoglobin). Xanthochromia results obtained with 4 different criteria (Chalmers original, Modified Chalmers, Duiser and UK NEQAS 2008) were compared. Results: We reviewed 41 medical charts (2 patients with duplicate lumbar punctures (LP) for a total of 43 LP). For these 41 patients there were 11 positive xanthochromia results, 5 of which were in concordance with a final diagnosis of subarachnoid hemorrhage (SAH). The diagnosis of the 6 other positive xanthochromia results were as follow: meningeal spread of a lymphoma, cerebral amyloid angiopathy, exertional headache, viral encephalitis with a possibility of petechiaes on the cerebral CT and second LP. Interpretation (negative/positive) of 40/43 LP was identical for the 4 methods. 2 LP were positive with Duiser and UK NEQAS 2008 but negative with Chalmers approaches (final diagnosis: SAH and cerebral amyloid angiopathy). 1 LP was positive only by the Duiser method (viral encephalitis). Conclusions: UK NEQAS 2008 guidelines identified all SAH but are sensitive to traumatic and pathologic meningeal lesions. Except for a case of viral encephalitis with a suspicion of cerebral petechiaes on CT, UK NEQAS 2008 gave xanthochromia results similar to the one in use at HEJ (Duiser). Chalmers original and Modified Chalmers methods missed one of the five SAH.

Cerebral Amyloid Angiopathy in Combination with Paroxysmal Atrial Fibrillation

2020 ◽  
Vol 25 (4) ◽  
pp. 31-37
Author(s):  
A. A. Kornilova ◽  
O. V. Lagoda ◽  
M. M. Tanashyan
Keyword(s):  
Atrial Fibrillation ◽  
Cerebral Amyloid Angiopathy ◽  
Paroxysmal Atrial Fibrillation ◽  
Past History ◽  
Laboratory Data ◽  
Clinical Manifestations ◽  
Cerebral Haemorrhage ◽  
Exclusion Criterion ◽  
Amyloid Angiopathy ◽  
Cerebral Amyloid

The present article addresses the definition of cerebral amyloid angiopathy (CAA) and its symptoms based on the analysis of the medical case; the issues of diagnosis and treatment of this pathology are discussed. The Boston criteria, which became the basis for diagnosis, study of clinical manifestations and progression of CAA and approaches to its therapy, are presented. Methods and modes of neuroimaging, including magnetic resonance imaging (MRI), which verify micro cerebral haemorrhage, are described. At the same time, the role and significance of cardiac arrhythmias in the genesis of ischemic stroke are discussed, and scales for assessing the risk of its occurrence are presented. The observation of the neurological, somatic, neuroimaging, neuropsychological status of a 62-year-old patient confirms quite rare combination of probable CAA, paroxysmal atrial fibrillation and repeated hemorrhagic functional apoplexy (FA). The relevance of the case described, is a complex clinical dilemma based on mutually exclusive recommendations for the pharmacological correction of such conditions. It is emphasized that in many multicenter clinical studies on the effectiveness of antithrombotic medication (antiaggregants, anticoagulants) in the treatment and prevention of ischaemic functional apoplexy , an important exclusion criterion is a hemorrhagic stroke in past history (including the multiple changes in haemostasis indicators). Taking into account the obtained clinical and laboratory data in the dynamics, the tactics of treating the described patient were determined. The results of studies related to the treatment of comorbid pathology that should become the subject of the development of a personalized algorithm for managing patients in each specific case, are discussed.

scholarly journals 10-Year Follow-Up of a Patient with Cerebral Amyloid Angiopathy

2020 ◽  
Vol 12 (Suppl. 1) ◽  
pp. 202-206
Author(s):  
Min Kyoung Kang ◽  
Byung-Woo Yoon
Keyword(s):  
Cerebral Amyloid Angiopathy ◽  
The Elderly ◽  
Amyloid Angiopathy ◽  
Spontaneous Intracerebral Hemorrhage ◽  
Radiological Features ◽  
Magnetic Imaging ◽  
Long Term Follow Up ◽  
Cerebral Amyloid

We report the case of long-term follow-up of brain magnetic imaging of cerebral amyloid angiopathy. Cerebral amyloid angiopathy is often considered a major cause of spontaneous intracerebral hemorrhage in the elderly. This case illustrates the markedly progressive clinical and radiological features of the vasculopathic process in 10 years.

scholarly journals Cerebral microbleed distribution following cardiac surgery can mimic cerebral amyloid angiopathy

2021 ◽  
Vol 3 (2) ◽  
pp. e000166
Author(s):  
Michele De Sciscio ◽  
Paul De Sciscio ◽  
Wilson Vallat ◽  
Timothy Kleinig
Keyword(s):  
Cardiac Surgery ◽  
Valve Replacement ◽  
Cerebral Amyloid Angiopathy ◽  
Artery Bypass ◽  
Area Under The Curve ◽  
Subcortical White Matter ◽  
Amyloid Angiopathy ◽  
Receiver Operator Curve Analysis ◽  
Cerebral Amyloid ◽  
Cabg Patients

Background and aimsHaving anecdotally noted a high frequency of lobar-restricted cerebral microbleeds (CMBs) mimicking cerebral amyloid angiopathy (CAA) in patients with previous cardiac surgery (especially valve replacement) presenting to our transient ischaemic attack (TIA) clinic, we set out to objectively determine the frequency and distribution of microbleeds in this population.MethodsWe performed a retrospective comparative cohort study in consecutive patients presenting to two TIA clinics with either: (1) previous coronary artery bypass grafting (CABG) (n=41); (2) previous valve replacement (n=41) or (3) probable CAA (n=41), as per the Modified Boston Criteria, without prior cardiac surgery. Microbleed number and distribution was determined and compared.ResultsAt least one lobar-restricted microbleed was found in the majority of cardiac surgery patients (65%) and 32/82 (39%) met diagnostic criteria for CAA. Valve replacement patients had a higher microbleed prevalence (90 vs 51%, p<0.01) and lobar-restricted microbleed count (2.6±2.7 vs 1.0±1.4, p<0.01) than post-CABG patients; lobar-restricted microbleed count in both groups was substantially less than in CAA patients (15.5±20.4, p<0.01). In postcardiac surgery patients, subcortical white matter (SWM) microbleeds were proportionally more frequent compared with CAA patients. Receiver operator curve analysis of a ‘location-based’ ratio (calculated as SWM/SWM+strictly-cortical CMBs), revealed an optimal ratio of 0.45 in distinguishing cardiac surgery-associated microbleeds from CAA (sensitivity 0.56, specificity 0.93, area under the curve 0.71).ConclusionLobar-restricted microbleeds are common in patients with past cardiac surgery, however a higher proportion of these CMBs involve the SWM than in patients with CAA.

Sign in / Sign up

Export Citation Format

Share Document