Associations between sex, body mass index and the individual microglial response in Alzheimer's disease.

BACKGROUND AND OBJECTIVES: 18-kDa translocator protein position-emission-tomography (TSPO-PET) imaging emerged for in vivo assessment of neuroinflammation in Alzheimer's disease (AD) research. Sex and obesity effects on TSPO-PET binding have been reported for cognitively normal humans (CN), but such effects have not yet been systematically evaluated in patients with AD. Thus, we aimed to investigate the impact of sex and obesity on the relationship between ß-amyloid-accumulation and microglial activation in AD. METHODS: 49 patients with AD (29 females, all Aß-positive) and 15 Aß-negative CN (8 female) underwent TSPO-PET ([18F]GE-180) and ß-amyloid-PET ([18F]flutemetamol) imaging. In 24 patients with AD (14 females), tau-PET ([18F]PI-2620) was additionally available. The brain was parcellated into 218 cortical regions and standardized-uptake-value-ratios (SUVr, cerebellar reference) were calculated. Per region and tracer, the regional increase of PET SUVr (z-score) was calculated for AD against CN. The regression derived linear effect of regional Aß-PET on TSPO-PET was used to determine the Aß-plaque-dependent microglial response (slope) and the Aß-plaque-independent microglial response (intercept) at the individual patient level. All read-outs were compared between sexes and tested for a moderation effect of sex on associations with body mass index (BMI). RESULTS: In AD, females showed higher mean cortical TSPO-PET z-scores (0.91 ± 0.49; males 0.30 ± 0.75; p = 0.002), while Aß-PET z-scores were similar. The Aß-plaque-independent microglial response was stronger in females with AD (+ 0.37 ± 0.38; males with AD - 0.33 ± 0.87; p = 0.006), pronounced at the prodromal stage. On the contrary, the Aß-plaque-dependent microglial response was not different between sexes. The Aß-plaque-independent microglial response was significantly associated with tau-PET in females (Braak-II regions: r = 0.757, p = 0.003), but not in males. BMI and the Aß-plaque-independent microglial response were significantly associated in females (r = 0.44, p = 0.018) but not in males (BMI*sex interaction: F(3,52) = 3.077, p = 0.005). CONCLUSION: While microglia response to fibrillar Aß is similar between sexes, women with AD show a stronger Aß-plaque-independent microglia response. This sex difference in Aß-independent microglial activation may be associated with tau accumulation. BMI is positively associated with the Aß-plaque-independent microglia response in females with AD but not in males, indicating that sex and obesity need to be considered when studying neuroinflammation in AD.

Individual regional associations between Aß-, tau- and neurodegeneration (ATN) with microglial activation in patients with primary and secondary tauopathies.

ß-amyloid (Aß) and tau aggregation as well as neuronal injury and atrophy (ATN) are the major hallmarks of Alzheimer's disease (AD), and biomarkers for these hallmarks have been linked to neuroinflammation. However, the detailed regional associations of these biomarkers with microglial activation in individual patients remain to be elucidated. We investigated a cohort of 55 patients with AD and primary tauopathies and 10 healthy controls that underwent TSPO-, Aß-, tau-, and perfusion-surrogate-PET, as well as structural MRI. Z-score deviations for 246 brain regions were calculated and biomarker contributions of Aß (A), tau (T), perfusion (N1), and gray matter atrophy (N2) to microglial activation (TSPO, I) were calculated for each individual subject. Individual ATN-related microglial activation was correlated with clinical performance and CSF soluble TREM2 (sTREM2) concentrations. In typical and atypical AD, regional tau was stronger and more frequently associated with microglial activation when compared to regional Aß (AD: ßT = 0.412 ± 0.196 vs. ßA = 0.142 ± 0.123, p < 0.001; AD-CBS: ßT = 0.385 ± 0.176 vs. ßA = 0.131 ± 0.186, p = 0.031). The strong association between regional tau and microglia reproduced well in primary tauopathies (ßT = 0.418 ± 0.154). Stronger individual associations between tau and microglial activation were associated with poorer clinical performance. In patients with 4RT, sTREM2 levels showed a positive association with tau-related microglial activation. Tau pathology has strong regional associations with microglial activation in primary and secondary tauopathies. Tau and Aß related microglial response indices may serve as a two-dimensional in vivo assessment of neuroinflammation in neurodegenerative diseases.

Microglial Activation and Connectivity in Alzheimer Disease and Aging.

OBJECTIVE: Alzheimer disease (AD) is characterized by amyloid ß (Aß) plaques and neurofibrillary tau tangles, but increasing evidence suggests that neuroinflammation also plays a key role, driven by the activation of microglia. Aß and tau pathology appear to spread along pathways of highly connected brain regions, but it remains elusive whether microglial activation follows a similar distribution pattern. Here, we assess whether connectivity is associated with microglia activation patterns. METHODS: We included 32 Aß-positive early AD subjects (18 women, 14 men) and 18 Aß-negative age-matched healthy controls (10 women, 8 men) from the prospective ActiGliA (Activity of Cerebral Networks, Amyloid and Microglia in Aging and Alzheimer's Disease) study. All participants underwent microglial activation positron emission tomography (PET) with the third-generation mitochondrial 18 kDa translocator protein (TSPO) ligand [18 F]GE-180 and magnetic resonance imaging (MRI) to measure resting-state functional and structural connectivity. RESULTS: We found that inter-regional covariance in TSPO-PET and standardized uptake value ratio was preferentially distributed along functionally highly connected brain regions, with MRI structural connectivity showing a weaker association with microglial activation. AD patients showed increased TSPO-PET tracer uptake bilaterally in the anterior medial temporal lobe compared to controls, and higher TSPO-PET uptake was associated with cognitive impairment and dementia severity in a disease stage-dependent manner. INTERPRETATION: Microglial activation distributes preferentially along highly connected brain regions, similar to tau pathology. These findings support the important role of microglia in neurodegeneration, and we speculate that pathology spreads throughout the brain along vulnerable connectivity pathways. ANN NEUROL 2022;92:768-781.

Pneumothorax detection in chest radiographs: optimizing artificial intelligence system for accuracy and confounding bias reduction using in-image annotations in algorithm training.

OBJECTIVES: Diagnostic accuracy of artificial intelligence (AI) pneumothorax (PTX) detection in chest radiographs (CXR) is limited by the noisy annotation quality of public training data and confounding thoracic tubes (TT). We hypothesize that in-image annotations of the dehiscent visceral pleura for algorithm training boosts algorithm's performance and suppresses confounders. METHODS: Our single-center evaluation cohort of 3062 supine CXRs includes 760 PTX-positive cases with radiological annotations of PTX size and inserted TTs. Three step-by-step improved algorithms (differing in algorithm architecture, training data from public datasets/clinical sites, and in-image annotations included in algorithm training) were characterized by area under the receiver operating characteristics (AUROC) in detailed subgroup analyses and referenced to the well-established "CheXNet" algorithm. RESULTS: Performances of established algorithms exclusively trained on publicly available data without in-image annotations are limited to AUROCs of 0.778 and strongly biased towards TTs that can completely eliminate algorithm's discriminative power in individual subgroups. Contrarily, our final "algorithm 2" which was trained on a lower number of images but additionally with in-image annotations of the dehiscent pleura achieved an overall AUROC of 0.877 for unilateral PTX detection with a significantly reduced TT-related confounding bias. CONCLUSIONS: We demonstrated strong limitations of an established PTX-detecting AI algorithm that can be significantly reduced by designing an AI system capable of learning to both classify and localize PTX. Our results are aimed at drawing attention to the necessity of high-quality in-image localization in training data to reduce the risks of unintentionally biasing the training process of pathology-detecting AI algorithms. KEY POINTS: • Established pneumothorax-detecting artificial intelligence algorithms trained on public training data are strongly limited and biased by confounding thoracic tubes. • We used high-quality in-image annotated training data to effectively boost algorithm performance and suppress the impact of confounding thoracic tubes. • Based on our results, we hypothesize that even hidden confounders might be effectively addressed by in-image annotations of pathology-related image features.

Clinically focused multi-cohort benchmarking as a tool for external validation of artificial intelligence algorithm performance in basic chest radiography analysis.

Artificial intelligence (AI) algorithms evaluating [supine] chest radiographs ([S]CXRs) have remarkably increased in number recently. Since training and validation are often performed on subsets of the same overall dataset, external validation is mandatory to reproduce results and reveal potential training errors. We applied a multicohort benchmarking to the publicly accessible (S)CXR analyzing AI algorithm CheXNet, comprising three clinically relevant study cohorts which differ in patient positioning ([S]CXRs), the applied reference standards (CT-/[S]CXR-based) and the possibility to also compare algorithm classification with different medical experts' reading performance. The study cohorts include [1] a cohort, characterized by 563 CXRs acquired in the emergency unit that were evaluated by 9 readers (radiologists and non-radiologists) in terms of 4 common pathologies, [2] a collection of 6,248 SCXRs annotated by radiologists in terms of pneumothorax presence, its size and presence of inserted thoracic tube material which allowed for subgroup and confounding bias analysis and [3] a cohort consisting of 166 patients with SCXRs that were evaluated by radiologists for underlying causes of basal lung opacities, all of those cases having been correlated to a timely acquired computed tomography scan (SCXR and CT within < 90 min). CheXNet non-significantly exceeded the radiology resident (RR) consensus in the detection of suspicious lung nodules (cohort [1], AUC AI/RR: 0.851/0.839, p = 0.793) and the radiological readers in the detection of basal pneumonia (cohort [3], AUC AI/reader consensus: 0.825/0.782, p = 0.390) and basal pleural effusion (cohort [3], AUC AI/reader consensus: 0.762/0.710, p = 0.336) in SCXR, partly with AUC values higher than originally published ("Nodule": 0.780, "Infiltration": 0.735, "Effusion": 0.864). The classifier "Infiltration" turned out to be very dependent on patient positioning (best in CXR, worst in SCXR). The pneumothorax SCXR cohort [2] revealed poor algorithm performance in CXRs without inserted thoracic material and in the detection of small pneumothoraces, which can be explained by a known systematic confounding error in the algorithm training process. The benefit of clinically relevant external validation is demonstrated by the differences in algorithm performance as compared to the original publication. Our multi-cohort benchmarking finally enables the consideration of confounders, different reference standards and patient positioning as well as the AI performance comparison with differentially qualified medical readers.

Impact of Confounding Thoracic Tubes and Pleural Dehiscence Extent on Artificial Intelligence Pneumothorax Detection in Chest Radiographs.

OBJECTIVES: We hypothesized that published performances of algorithms for artificial intelligence (AI) pneumothorax (PTX) detection in chest radiographs (CXRs) do not sufficiently consider the influence of PTX size and confounding effects caused by thoracic tubes (TTs). Therefore, we established a radiologically annotated benchmarking cohort (n = 6446) allowing for a detailed subgroup analysis. MATERIALS AND METHODS: We retrospectively identified 6434 supine CXRs, among them 1652 PTX-positive cases and 4782 PTX-negative cases. Supine CXRs were radiologically annotated for PTX size, PTX location, and inserted TTs. The diagnostic performances of 2 AI algorithms ("AI_CheXNet" [Rajpurkar et al], "AI_1.5" [Guendel et al]), both trained on publicly available datasets with labels obtained from automatic report interpretation, were quantified. The algorithms' discriminative power for PTX detection was quantified by the area under the receiver operating characteristics (AUROC), and significance analysis was based on the corresponding 95% confidence interval. A detailed subgroup analysis was performed to quantify the influence of PTX size and the confounding effects caused by inserted TTs. RESULTS: Algorithm performance was quantified as follows: overall performance with AUROCs of 0.704 (AI_1.5) / 0.765 (AI_CheXNet) for unilateral PTXs, AUROCs of 0.666 (AI_1.5) / 0.722 (AI_CheXNet) for unilateral PTXs smaller than 1 cm, and AUROCs of 0.735 (AI_1.5) / 0.818 (AI_CheXNet) for unilateral PTXs larger than 2 cm. Subgroup analysis identified TTs to be strong confounders that significantly influence algorithm performance: Discriminative power is completely eliminated by analyzing PTX-positive cases without TTs referenced to control PTX-negative cases with inserted TTs. Contrarily, AUROCs increased up to 0.875 (AI_CheXNet) for large PTX-positive cases with inserted TTs referenced to control cases without TTs. CONCLUSIONS: Our detailed subgroup analysis demonstrated that the performance of established AI algorithms for PTX detection trained on public datasets strongly depends on PTX size and is significantly biased by confounding image features, such as inserted TTS. Our established, clinically relevant and radiologically annotated benchmarking cohort might be of great benefit for ongoing algorithm development.