Differential diagnosis of neurodegenerative dementias with the explainable MRI based machine learning algorithm MUQUBIA.

Biomarker-based differential diagnosis of the most common forms of dementia is becoming increasingly important. Machine learning (ML) may be able to address this challenge. The aim of this study was to develop and interpret a ML algorithm capable of differentiating Alzheimer's dementia, frontotemporal dementia, dementia with Lewy bodies and cognitively normal control subjects based on sociodemographic, clinical, and magnetic resonance imaging (MRI) variables. 506 subjects from 5 databases were included. MRI images were processed with FreeSurfer, LPA, and TRACULA to obtain brain volumes and thicknesses, white matter lesions and diffusion metrics. MRI metrics were used in conjunction with clinical and demographic data to perform differential diagnosis based on a Support Vector Machine model called MUQUBIA (Multimodal Quantification of Brain whIte matter biomArkers). Age, gender, Clinical Dementia Rating (CDR) Dementia Staging Instrument, and 19 imaging features formed the best set of discriminative features. The predictive model performed with an overall Area Under the Curve of 98%, high overall precision (88%), recall (88%), and F1 scores (88%) in the test group, and good Label Ranking Average Precision score (0.95) in a subset of neuropathologically assessed patients. The results of MUQUBIA were explained by the SHapley Additive exPlanations (SHAP) method. The MUQUBIA algorithm successfully classified various dementias with good performance using cost-effective clinical and MRI information, and with independent validation, has the potential to assist physicians in their clinical diagnosis.

Coexistence of Amyotrophic Lateral Sclerosis and Alzheimer's Disease: Case Report and Review of the Literature.

We describe a case of amyotrophic lateral sclerosis (ALS) associated with Alzheimer's disease (AD) and review the literature about the coexistence of the two entities, highlighting the following: mean age at onset is 63.8 years, with slight female predominance; ALS tends to manifest after cognitive impairment and often begins in the bulbar region; average disease duration is 3 years; cognitive phenotype is mostly amnestic; the pattern of brain involvement is, in most cases, consistent with AD. Our case and the reviewed ones suggest that patients with ALS and dementia lacking unequivocal features of FTD should undergo additional examinations in order to recognize AD.

Educational interventions to improve detection and management of cognitive decline in primary care-An Italian multicenter pragmatic study.

Introduction: Timely detection of cognitive decline in primary care is essential to promote an appropriate care pathway and enhance the benefits of interventions. We present the results of a study aimed to evaluate the effectiveness of an educational intervention addressed to Italian family physicians (FPs) to improve timely detection and management of cognitive decline. Materials and methods: We conducted a pre-post study in six Italian health authorities (HAs) involving 254 FPs and 3,736 patients. We measured process and outcome indicators before the intervention (1 January 2014 to 31 December 2016) and after the intervention (1 January 2018 to 31 December 2019). One interactive face-to-face session workshop was delivered by local cognitive disorders and dementia specialists and FP advisors at each HA, in the period September 2017-December 2017. The session focused on key messages of the local Diagnostic and Therapeutic Care Pathway (DTCP) or regional guidelines: (a) the role of the FP for a timely suspicion of cognitive decline is fundamental; (b) when cognitive decline is suspected, the role of the FP is active in the diagnostic work-up; (c) FP's knowledge on pharmacological and non-pharmacological interventions is essential to improve the management of patients with cognitive decline. Results: An overall improvement in diagnostic procedures and management of patients with cognitive decline by FPs after the intervention was observed. The number of visits per year performed by FPs increased, and the time interval between the first FP consultation and the diagnosis was optimized. Neuroleptic use significantly decreased, whereas the use of benzodiazepines remained steadily high. Non-pharmacological interventions, or use of support services, were underrepresented even in the post-intervention. Differences among the participating HAs were identified and discussed. Discussion: Results from this study suggest the success of the educational intervention addressed to FPs in improving early detection and management of cognitive decline, highlighting the importance to continue medical education in this field. At the same time, further initiatives of care pathway dissemination and implementation should promote strategies to enhance interactions between primary and secondary care optimizing the collaboration between FPs and specialists.

Italian adaptation of the Uniform Data Set Neuropsychological Test Battery (I-UDSNB 1.0): development and normative data.

BACKGROUND: Neuropsychological testing plays a cardinal role in the diagnosis and monitoring of Alzheimer's disease. A major concern is represented by the heterogeneity of the neuropsychological batteries currently adopted in memory clinics and healthcare centers. The current study aimed to solve this issue. METHODS: Following the initiative of the University of Washington's National Alzheimer's Coordinating Center (NACC), we presented the Italian adaptation of the Neuropsychological Test Battery of the Uniform Data Set (I-UDSNB). We collected data from 433 healthy Italian individuals and employed regression models to evaluate the impact of demographic variables on the performance, deriving the reference norms. RESULTS: Higher education and lower age were associated with a better performance in the majority of tests, while sex affected only fluency tests and Digit Span Forward. CONCLUSIONS: The I-UDSNB offers a valuable and harmonized tool for neuropsychological testing in Italy, to be used in clinical and research settings.

Brain Metabolism and Amyloid Load in Individuals With Subjective Cognitive Decline or Pre-Mild Cognitive Impairment.

BACKGROUND AND OBJECTIVE: Multicenter study aiming at investigating the characteristics of cognitive decline, neuropsychiatric symptoms, and brain imaging in individuals with subjective cognitive decline (SCD) and subtle cognitive decline (pre-Mild Cognitive Impairment, pre-MCI). METHODS: Data were obtained from the Network-AD project (NET-2011-02346784). The included subjects underwent baseline cognitive and neurobehavioral evaluation, FDG-PET, and, amyloid-PET. We used Principal Component Analysis (PCA) to identify independent neuropsychological and neuropsychiatric dimensions and their association with brain metabolism. RESULTS: A total of 105 subjects (SCD=49, pre-MCI=56) were included. FDG-PET was normal in 45% of subjects and revealed brain hypometabolism in 55%, with a frontal-like pattern as the most frequent finding (28%). Neuropsychiatric symptoms emerging from the Neuropsychiatric Inventory and the Starkstein Apathy Scale were highly prevalent in the whole sample (78%). An abnormal amyloid load was detected in the 18% of the subjects that underwent amyloid-PET (n=60). PCA resulted in three neuropsychological factors: 1) executive/visuo-motor, correlating with hypometabolism in frontal, occipital cortices and basal ganglia; 2) memory, correlating with hypometabolism in temporo-parietal regions; 3) visuo-spatial/constructional, correlating with hypometabolism in fronto-parietal cortices. Two factors emerged from the neuropsychiatric PCA: 1) affective, correlating with hypometabolism in orbito-frontal, cingulate cortex, insula; 2) hyperactive/psychotic, correlating with hypometabolism in frontal, temporal and parietal regions. DISCUSSION: FDG-PET evidence suggests either normal brain function or different patterns of brain hypometabolism in SCD and pre-MCI subjects. These results indicate that SCD and pre-MCI represent heterogeneous populations. Consistently, different neuropsychological and neuropsychiatric profiles emerged, which correlated with neuronal dysfunction in specific brain regions. Long-term follow-up studies are needed to assess the risk of progression to dementia in these conditions.

Accuracy of the clinical diagnosis of dementia with Lewy bodies (DLB) among the Italian Dementia Centers: a study by the Italian DLB study group (DLB-SINdem).

INTRODUCTION: Dementia with Lewy bodies (DLB) may represent a diagnostic challenge, since its clinical picture overlaps with other dementia. Two toolkits have been developed to aid the clinician to diagnose DLB: the Lewy Body Composite Risk Score (LBCRS) and the Assessment Toolkit for DLB (AT-DLB). We aim to evaluate the reliability of these two questionnaires, and their ability to enhance the interpretation of the international consensus diagnostic criteria. METHODS: LBCRS and AT-DLB were distributed to 135 Italian Neurological Centers for Cognitive Decline and Dementia (CDCDs), with the indication to administer them to all patients with dementia referred within the subsequent 3 months. We asked to subsequently apply consensus criteria for DLB diagnosis, to validate the diagnostic accuracy of the two toolkits. RESULTS: A total of 23 Centers joined the study; 1854 patients were enrolled. We found a prevalence of possible or probable DLB of 13% each (26% total), according to the consensus criteria. LBCRS toolkit showed good reliability, with a Cronbach alpha of 0.77, stable even after removing variables from the construct. AT-DLB toolkit Cronbach alpha was 0.52 and, after the subtraction of the "cognitive fluctuation" criterion, was only 0.31. Accuracy, sensitivity, and specificity were higher for LBCRS vs. AT-DLB. However, when simultaneously considered in the logistic models, AT-DLB showed a better performance (p < 0.001). Overall, the concordance between LBCRS positive and AT-DLB possible/probable was of 78.02% CONCLUSIONS: In a clinical setting, the LBCRS and AT-DLB questionnaires have good accuracy for DLB diagnosis.

Psychological Impact of Predictive Genetic Testing for Inherited Alzheimer Disease and Frontotemporal Dementia: The IT-DIAfN Protocol.

AIM: Our aim was to evaluate the psychological impact of predictive genetic testing in individuals at-risk for inherited dementia who underwent a structured counseling and testing protocol. METHODS: Participants were healthy at-risk relatives from families with at least one affected patient, in whom a disease-associated genetic variant had been ascertained. A comprehensive psychological assessment (personality, anxiety and depression, quality of life, coping strategies, resilience and health-related beliefs) was administered at baseline, at 6 months and 12 months follow-up. RESULTS: Twenty-four participants from 13 families were included. Sixteen participants underwent blood sampling and genetic analysis; 6 resulted to be carriers of pathogenic variants (1 in PSEN1, 1 in PSEN2, 4 in GRN). Carriers showed higher score on the Resilience Scale for Adults (RSA) - social competence, and on Multidimensional Health Locus of Control - internal, than noncarriers (P=0.03 for both). Ten at-risk relatives who completed the follow-up showed improvement in RSA - planned future (P=0.01) with respect to baseline. DISCUSSION: Our case series showed that at-risk individuals undergoing predictive testing showed benefit on personal life and no detrimental impact on a broad range of psychological outcomes. Higher social skills and lower internal health locus of control in carriers may be an early psychological correlate of preclinical dementia.

Postinfectious Neurologic Complications in COVID-19: A Complex Case Report.

A 40-y-old woman with severe acute respiratory syndrome coronavirus 2 infection developed neurologic manifestations (confusion, agitation, seizures, dyskinesias, and parkinsonism) a few weeks after the onset of severe acute respiratory syndrome. MRI and cerebrospinal fluid analyses were unremarkable, but 18F-FDG PET/CT showed limbic and extralimbic hypermetabolism. A full recovery, alongside 18F-FDG normalization in previously hypermetabolic areas, was observed after intravenous immunoglobulin administration.

Neuro-telehealth for fragile patients in a tertiary referral neurological institute during the COVID-19 pandemic in Milan, Lombardy.

BACKGROUND: Lombardy was severely hit by the COVID-19 pandemic since February 2020 and the Health System underwent rapid reorganization. Outpatient clinics were stopped for non-urgent patients: it became a priority to manage hundreds of fragile neurological patients who suddenly had less reference points. In Italy, before the pandemic, Televisits were neither recognized nor priced. METHODS: At the Fondazione IRCCS Istituto Neurologico C. Besta, we reorganized outpatient clinics to deliver Neuro-telemedicine services, including Televisits and Teleneurorehabilitation, since March 2020. A dedicated Working Group prepared the procedure, tested the system, and designed satisfaction questionnaires for adults and children. RESULTS: After a pilot phase, we prepared a procedure for Telemedicine outpatient clinics which was approved by hospital directions. It included prescription, booking, consenting, privacy and data protection, secure connection with patients (Teams Microsoft 365), electronic report preparation and delivery, reporting, and accountability of the services. During the March-September 2020 period, we delivered 3167 Telemedicine services, including 1618 Televisits, to 1694 patients (972 adults, 722 children) with a wide range of chronic neurological disorders. We successfully administered different clinical assessment and scales. Satisfaction among patients and caregivers was very high. CONCLUSIONS: During the dramatic emergency, we were able to take care of more than 1600 patients by organizing Neuro-telehealth in a few weeks, lessening the impact of the pandemic on fragile patients with chronic neurological disorders; this strategy is now stably embedded in our care pathways. In Italy, Telehealth is at present recognized and priced and is becoming a stable pillar of the health system.

Automatic multispectral MRI segmentation of human hippocampal subfields: an evaluation of multicentric test-retest reproducibility.

Accurate and reproducible automated segmentation of human hippocampal subfields is of interest to study their roles in cognitive functions and disease processes. Multispectral structural MRI methods have been proposed to improve automated hippocampal subfield segmentation accuracy, but the reproducibility in a multicentric setting is, to date, not well characterized. Here, we assessed test-retest reproducibility of FreeSurfer 6.0 hippocampal subfield segmentations using multispectral MRI analysis pipelines (22 healthy subjects scanned twice, a week apart, at four 3T MRI sites). The harmonized MRI protocol included two 3D-T1, a 3D-FLAIR, and a high-resolution 2D-T2. After within-session T1 averaging, subfield volumes were segmented using three pipelines with different multispectral data: two longitudinal ("long_T1s" and "long_T1s_FLAIR") and one cross-sectional ("long_T1s_FLAIR_crossT2"). Volume reproducibility was quantified in magnitude (reproducibility error-RE) and space (DICE coefficient). RE was lower in all hippocampal subfields, except for hippocampal fissure, using the longitudinal pipelines compared to long_T1s_FLAIR_crossT2 (average RE reduction of 0.4-3.6%). Similarly, the longitudinal pipelines showed a higher spatial reproducibility (1.1-7.8% of DICE improvement) in all hippocampal structures compared to long_T1s_FLAIR_crossT2. Moreover, long_T1s_FLAIR provided a small but significant RE improvement in comparison to long_T1s (p = 0.015), whereas no significant DICE differences were found. In addition, structures with volumes larger than 200 mm3 had better RE (1-2%) and DICE (0.7-0.95) than smaller structures. In summary, our study suggests that the most reproducible hippocampal subfield FreeSurfer segmentations are derived from a longitudinal pipeline using 3D-T1s and 3D-FLAIR. Adapting a longitudinal pipeline to include high-resolution 2D-T2 may lead to further improvements.

Medical Informatics Platform (MIP): A Pilot Study Across Clinical Italian Cohorts.

Introduction: With the shift of research focus to personalized medicine in Alzheimer's Dementia (AD), there is an urgent need for tools that are capable of quantifying a patient's risk using diagnostic biomarkers. The Medical Informatics Platform (MIP) is a distributed e-infrastructure federating large amounts of data coupled with machine-learning (ML) algorithms and statistical models to define the biological signature of the disease. The present study assessed (i) the accuracy of two ML algorithms, i.e., supervised Gradient Boosting (GB) and semi-unsupervised 3C strategy (Categorize, Cluster, Classify-CCC) implemented in the MIP and (ii) their contribution over the standard diagnostic workup. Methods: We examined individuals coming from the MIP installed across 3 Italian memory clinics, including subjects with Normal Cognition (CN, n = 432), Mild Cognitive Impairment (MCI, n = 456), and AD (n = 451). The GB classifier was applied to best discriminate the three diagnostic classes in 1,339 subjects, and the CCC strategy was used to refine the classical disease categories. Four dementia experts provided their diagnostic confidence (DC) of MCI conversion on an independent cohort of 38 patients. DC was based on clinical, neuropsychological, CSF, and structural MRI information and again with addition of the outcome from the MIP tools. Results: The GB algorithm provided a classification accuracy of 85% in a nested 10-fold cross-validation for CN vs. MCI vs. AD discrimination. Accuracy increased to 95% in the holdout validation, with the omission of each Italian clinical cohort out in turn. CCC identified five homogeneous clusters of subjects and 36 biomarkers that represented the disease fingerprint. In the DC assessment, CCC defined six clusters in the MCI population used to train the algorithm and 29 biomarkers to improve patients staging. GB and CCC showed a significant impact, evaluated as +5.99% of increment on physicians' DC. The influence of MIP on DC was rated from "slight" to "significant" in 80% of the cases. Discussion: GB provided fair results in classification of CN, MCI, and AD. CCC identified homogeneous and promising classes of subjects via its semi-unsupervised approach. We measured the effect of the MIP on the physician's DC. Our results pave the way for the establishment of a new paradigm for ML discrimination of patients who will or will not convert to AD, a clinical priority for neurology.

Research criteria for the diagnosis of prodromal dementia with Lewy bodies.

The prodromal phase of dementia with Lewy bodies (DLB) includes (1) mild cognitive impairment (MCI), (2) delirium-onset, and (3) psychiatric-onset presentations. The purpose of our review is to determine whether there is sufficient information yet available to justify development of diagnostic criteria for each of these. Our goal is to achieve evidence-based recommendations for the recognition of DLB at a predementia, symptomatic stage. We propose operationalized diagnostic criteria for probable and possible mild cognitive impairment with Lewy bodies, which are intended for use in research settings pending validation for use in clinical practice. They are compatible with current criteria for other prodromal neurodegenerative disorders including Alzheimer and Parkinson disease. Although there is still insufficient evidence to propose formal criteria for delirium-onset and psychiatric-onset presentations of DLB, we feel that it is important to characterize them, raising the index of diagnostic suspicion and prioritizing them for further investigation.

Keeping People with Dementia or Mild Cognitive Impairment in Employment: A Literature Review on Its Determinants.

Background: Approximately 10-20% of people with early onset dementias (EOD) or mild cognitive impairment (MCI) are aged under 65 and, due to extended working life and increasing prevalence of dementias, they more and more frequently will be present in the active workforce. This review aimed to synthesize the available information about the ability of people with EOD or MCI to retain their participation in the labor workforce. Methods: We searched SCOPUS and EMBASE for peer-reviewed papers that reported studies assessing work ability in employees with EOD or MCI that were published in the period of January 2010 to August 2019. Results: We selected four publications, in which 1012 participants with EOD or MCI were enrolled (41.2% males). Cognitive difficulties rather than motor dysfunction were found to reduce patients' ability to work. Two main themes emerged: management of dementia in the workplace and the impact of symptoms on working status. Conclusions: EOD and MCI impact on workforce participation by determining problems in executive functions. Although this review was based on a small sample of studies, it can be shown that support in the workplace may act as a facilitator to enhance workforce participation, and occupational health professionals can help patients with EOD or MCI continue working as much as possible.

Comparison of visual criteria for amyloid-PET reading: could criteria merging reduce inter-rater variability?

BACKGROUND: Three different amyloid tracers labeled with 18-flourine have been introduced into clinical use. The leaflets of tracers indicate different visual criteria for PET reporting. In clinical practice, it is not yet ascertained whether these criteria are equivalent in terms of diagnostic accuracy or if anyone is better than another. We aimed to evaluate the inter and intra-rater variability of visual assessment of 18F-Florbetapir PET/CT images among six independent readers with different clinical experience. METHODS: We analyzed 252 PET/CT scans, visually assessed by each reader three times, applying independently the three different reading criteria proposed. Each reader evaluated the regional uptake specifying for each cortical region a numeric value of grading of positivity in order to assign a final score. At the end of each reading a level of confidence was determined by assigning a score from 0 (negative) to 4 (positive). After first reading, those cases in which the evaluations by two experienced readers did not match (discordant cases) were independently reevaluated merging all the three different visual interpretation criteria. RESULTS: Good agreement was observed for visual interpretation among the six readers' confidence-level using independently the three visual reading criteria: ICC=0.83 (0.80-0.86) for 18F-florbetapir, ICC=0.84 (0.81-0.87) for 18F-florbetaben, and ICC=0.86 (0.83-0.88) for 18F-flutemetamol reading. A good inter-rater agreement was observed for final-score too: ICC=0.74 (0.70-0.78) for 18F-florbetapir; ICC=0.82 (0.79-0.85) for 18F-florbetaben; ICC=0.84 (0.81-0.87) for 18F-flutemetamol. Intra-rater agreement was good for final-score (from 0.76 to 0.90; P<0.001) and confidence-level (Spearman's rho from 0.89 to 1.00; P<0.001). Disagreement between the two experienced readers was observed in 22 of 252 cases (9%). The agreement converged over a second round of independent reading in 12 of 22 cases (54%), by merging all the criteria. CONCLUSIONS: All the criteria proposed are useful to determine the grading of positivity or negativity of amyloid deposition and their merging improves the diagnostic confidence and provides a better agreement.

Low-dose CT for the spatial normalization of PET images: A validation procedure for amyloid-PET semi-quantification.

The reference standard for spatial normalization of brain positron emission tomography (PET) images involves structural Magnetic Resonance Imaging (MRI) data. However, the lack of such structural information is fairly common in clinical settings. This might lead to lack of proper image quantification and to evaluation based only on visual ratings, which does not allow research studies or clinical trials based on quantification. PET/CT systems are widely available and CT normalization procedures need to be explored. Here we describe and validate a procedure for the spatial normalization of PET images based on the low-dose Computed Tomography (CT) images contextually acquired for attenuation correction in PET/CT systems. We included N = 34 subjects, spanning from cognitively normal to mild cognitive impairment and dementia, who underwent amyloid-PET/CT (18F-Florbetaben) and structural MRI scans. The proposed pipeline is based on the SPM12 unified segmentation algorithm applied to low-dose CT images. The validation of the normalization pipeline focused on 1) statistical comparisons between regional and global 18F-Florbetaben-PET/CT standardized uptake value ratios (SUVrs) estimated from both CT-based and MRI-based normalized PET images (SUVrCT, SUVrMRI) and 2) estimation of the degrees of overlap between warped gray matter (GM) segmented maps derived from CT- and MRI-based spatial transformations. We found negligible deviations between regional and global SUVrs in the two CT and MRI-based methods. SUVrCT and SUVrMRI global uptake scores showed negligible differences (mean ±â€¯sd 0.01 ±â€¯0.03). Notably, the CT- and MRI-based warped GM maps showed excellent overlap (90% within 1 mm). The proposed analysis pipeline, based on low-dose CT images, allows accurate spatial normalization and subsequent PET image quantification. A CT-based analytical pipeline could benefit both research and clinical practice, allowing the recruitment of larger samples and favoring clinical routine analysis.

Quantitative appraisal of the Amyloid Imaging Taskforce appropriate use criteria for amyloid-PET.

INTRODUCTION: We test the hypothesis that amyloid-positron emission tomography prescriptions, considered appropriate based on the Amyloid Imaging Taskforce (AIT) criteria, lead to greater clinical utility than AIT-inappropriate prescriptions. METHODS: We compared the clinical utility between patients who underwent amyloid-positron emission tomography appropriately or inappropriately and among the subgroups of patients defined by the AIT criteria. Finally, we performed logistic regressions to identify variables associated with clinical utility. RESULTS: We identified 171 AIT-appropriate and 67 AIT-inappropriate patients. AIT-appropriate and AIT-inappropriate cases did not differ in any outcomes of clinical utility (P > .05). Subgroup analysis denoted both expected and unexpected results. The logistic regressions outlined the primary role of clinical picture and clinical or neuropsychological profile in identifying patients benefitting from amyloid-positron emission tomography. DISCUSSION: Contrary to our hypothesis, also AIT-inappropriate prescriptions were associated with clinical utility. Clinical or neuropsychological variables, not taken into account by the AIT criteria, may help further refine criteria for appropriateness.