Amyloid-ß prediction machine learning model using source-based morphometry across neurocognitive disorders.

Previous studies have developed and explored magnetic resonance imaging (MRI)-based machine learning models for predicting Alzheimer's disease (AD). However, limited research has focused on models incorporating diverse patient populations. This study aimed to build a clinically useful prediction model for amyloid-beta (Aß) deposition using source-based morphometry, using a data-driven algorithm based on independent component analyses. Additionally, we assessed how the predictive accuracies varied with the feature combinations. Data from 118 participants clinically diagnosed with various conditions such as AD, mild cognitive impairment, frontotemporal lobar degeneration, corticobasal syndrome, progressive supranuclear palsy, and psychiatric disorders, as well as healthy controls were used for the development of the model. We used structural MR images, cognitive test results, and apolipoprotein E status for feature selection. Three-dimensional T1-weighted images were preprocessed into voxel-based gray matter images and then subjected to source-based morphometry. We used a support vector machine as a classifier. We applied SHapley Additive exPlanations, a game-theoretical approach, to ensure model accountability. The final model that was based on MR-images, cognitive test results, and apolipoprotein E status yielded 89.8% accuracy and a receiver operating characteristic curve of 0.888. The model based on MR-images alone showed 84.7% accuracy. Aß-positivity was correctly detected in non-AD patients. One of the seven independent components derived from source-based morphometry was considered to represent an AD-related gray matter volume pattern and showed the strongest impact on the model output. Aß-positivity across neurological and psychiatric disorders was predicted with moderate-to-high accuracy and was associated with a probable AD-related gray matter volume pattern. An MRI-based data-driven machine learning approach can be beneficial as a diagnostic aid.

Performance of plasma Aß42/40, measured using a fully automated immunoassay, across a broad patient population in identifying amyloid status.

BACKGROUND: Plasma biomarkers have emerged as promising screening tools for Alzheimer's disease (AD) because of their potential to detect amyloid ß (Aß) accumulation in the brain. One such candidate is the plasma Aß42/40 ratio (Aß42/40). Unlike previous research that used traditional immunoassay, recent studies that measured plasma Aß42/40 using fully automated platforms reported promising results. However, its utility should be confirmed using a broader patient population, focusing on the potential for early detection. METHODS: We recruited 174 participants, including healthy controls (HC) and patients with clinical diagnoses of AD, frontotemporal lobar degeneration, dementia with Lewy bodies/Parkinson's disease, mild cognitive impairment (MCI), and others, from a university memory clinic. We examined the performance of plasma Aß42/40, measured using the fully automated high-sensitivity chemiluminescence enzyme (HISCL) immunoassay, in detecting amyloid-positron emission tomography (PET)-derived Aß pathology. We also compared its performance with that of Simoa-based plasma phosphorylated tau at residue 181 (p-tau181), glial fibrillary acidic protein (GFAP), and neurofilament light (NfL). RESULTS: Using the best cut-off derived from the Youden Index, plasma Aß42/40 yielded an area under the receiver operating characteristic curve (AUC) of 0.949 in distinguishing visually assessed 18F-Florbetaben amyloid PET positivity. The plasma Aß42/40 had a significantly superior AUC than p-tau181, GFAP, and NfL in the 167 participants with measurements for all four biomarkers. Next, we analyzed 99 participants, including only the HC and those with MCI, and discovered that plasma Aß42/40 outperformed the other plasma biomarkers, suggesting its ability to detect early amyloid accumulation. Using the Centiloid scale (CL), Spearman's rank correlation coefficient between plasma Aß42/40 and CL was -0.767. Among the 15 participants falling within the CL values indicative of potential future amyloid accumulation (CL between 13.5 and 35.7), plasma Aß42/40 categorized 61.5% (8/13) as Aß-positive, whereas visual assessment of amyloid PET identified 20% (3/15) as positive. CONCLUSION: Plasma Aß42/40 measured using the fully automated HISCL platform showed excellent performance in identifying Aß accumulation in the brain in a well-characterized cohort. This equipment may be useful for screening amyloid pathology because it has the potential to detect early amyloid pathology and is readily applied in clinical settings.

Language patterns in Japanese patients with Alzheimer disease: A machine learning approach.

AIM: The authors applied natural language processing and machine learning to explore the disease-related language patterns that warrant objective measures for assessing language ability in Japanese patients with Alzheimer disease (AD), while most previous studies have used large publicly available data sets in Euro-American languages. METHODS: The authors obtained 276 speech samples from 42 patients with AD and 52 healthy controls, aged 50 years or older. A natural language processing library for Python was used, spaCy, with an add-on library, GiNZA, which is a Japanese parser based on Universal Dependencies designed to facilitate multilingual parser development. The authors used eXtreme Gradient Boosting for our classification algorithm. Each unit of part-of-speech and dependency was tagged and counted to create features such as tag-frequency and tag-to-tag transition-frequency. Each feature's importance was computed during the 100-fold repeated random subsampling validation and averaged. RESULTS: The model resulted in an accuracy of 0.84 (SD = 0.06), and an area under the curve of 0.90 (SD = 0.03). Among the features that were important for such predictions, seven of the top 10 features were related to part-of-speech, while the remaining three were related to dependency. A box plot analysis demonstrated that the appearance rates of content words-related features were lower among the patients, whereas those with stagnation-related features were higher. CONCLUSION: The current study demonstrated a promising level of accuracy for predicting AD and found the language patterns corresponding to the type of lexical-semantic decline known as 'empty speech', which is regarded as a characteristic of AD.

Impact of Amyloid and Tau PET on Changes in Diagnosis and Patient Management.

BACKGROUND AND OBJECTIVES: Previous studies have evaluated the diagnostic effect of amyloid PET in selected research cohorts. However, these studies did not assess the clinical impact of the combination of amyloid and tau PETs. Our objective was to evaluate the association of the combination of 2 PETs with changes in diagnosis, treatment, and management in a memory clinic cohort. METHODS: All participants underwent amyloid [18F]florbetaben PET and tau PET using [18F]PI-2620 or [18F]Florzolotau, which are potentially useful for the diagnosis of non-Alzheimer disease (AD) tauopathies. Dementia specialists determined a pre- and post-PET diagnosis that existed in both a clinical syndrome (cognitive normal [CN], mild cognitive impairment [MCI], and dementia) and suspected etiology, with a confidence level. In addition, the dementia specialists determined patient treatment in terms of ancillary investigations and management. RESULTS: Among 126 registered participants, 84.9% completed the study procedures and were included in the analysis (CN [n = 40], MCI [n = 25], AD [n = 20], and non-AD dementia [n = 22]). The etiologic diagnosis changed in 25.0% in the CN, 68.0% in the MCI, and 23.8% with dementia. Overall changes in management between pre- and post-PET occurred in 5.0% of CN, 52.0% of MCI, and 38.1% of dementia. Logistic regression analysis revealed that tau PET has stronger associations with change management than amyloid PET in all participants and dementia groups. DISCUSSION: The combination of amyloid and tau PETs was associated with changes in management and diagnosis of MCI and dementia, and the second-generation tau PET has a strong impact on the changes in diagnosis and management in memory clinics. CLASSIFICATION OF EVIDENCE: This study provides Class I evidence that the combination of amyloid and tau PETs was associated with changes in management and diagnosis of MCI and dementia.

Case report: Non-Alzheimer's disease tauopathy with logopenic variant primary progressive aphasia diagnosed using amyloid and tau PET.

We report a patient with logopenic variant primary progressive aphasia (lv-PPA) who was diagnosed as having non-Alzheimer's disease (AD) tauopathy after multiple biophysical/biological examinations, including amyloid and 18F-florzolotau tau positron emission tomography (PET), had been performed. A woman in her late 60s who had previously been diagnosed as having AD was referred to us for a further, detailed examination. She had been unaware of any symptoms at the time of AD diagnosis, but she subsequently became gradually aware of a speech impairment. She talked nearly completely and fluently, although she occasionally exhibited word-finding difficulty and made phonological errors during naming, word fluency testing, and sentence repetition; these findings met the criteria for the diagnosis of lv-PPA, which is known to be observed more commonly in AD than in other proteinopathies. Magnetic resonance imaging, single photon emission computed tomography, and plasma phosphorylated tau and plasma neurofilament light chain measurements showed an AD-like pattern. However, both 11C-Pittsburgh compound-B and 18F-florbetaben amyloid PET showed negative results, whereas 18F-florzolotau tau PET yielded positive results, with radio signals predominantly in the left superior temporal gyrus, middle temporal gyrus, supramarginal gyrus, and frontal operculum. Whole-genome sequencing revealed no known dominantly inherited mutations in AD or frontotemporal lobar degeneration genes, including the genes encoding amyloid precursor protein, microtubule-associated protein tau, presenilin 1 and 2. To the best of our knowledge, this patient was a rare case of lv-PPA who was diagnosed as having non-AD tauopathy based on the results of multiple examinations, including whole-genome sequencing, plasma measurement, and amyloid and 18F-florzolotau tau PET. This case underscores the clinicopathologically heterogeneous nature of this syndrome.

Understanding psychiatric illness through natural language processing (UNDERPIN): Rationale, design, and methodology.

Introduction: Psychiatric disorders are diagnosed through observations of psychiatrists according to diagnostic criteria such as the DSM-5. Such observations, however, are mainly based on each psychiatrist's level of experience and often lack objectivity, potentially leading to disagreements among psychiatrists. In contrast, specific linguistic features can be observed in some psychiatric disorders, such as a loosening of associations in schizophrenia. Some studies explored biomarkers, but biomarkers have yet to be used in clinical practice. Aim: The purposes of this study are to create a large dataset of Japanese speech data labeled with detailed information on psychiatric disorders and neurocognitive disorders to quantify the linguistic features of those disorders using natural language processing and, finally, to develop objective and easy-to-use biomarkers for diagnosing and assessing the severity of them. Methods: This study will have a multi-center prospective design. The DSM-5 or ICD-11 criteria for major depressive disorder, bipolar disorder, schizophrenia, and anxiety disorder and for major and minor neurocognitive disorders will be regarded as the inclusion criteria for the psychiatric disorder samples. For the healthy subjects, the absence of a history of psychiatric disorders will be confirmed using the Mini-International Neuropsychiatric Interview (M.I.N.I.). The absence of current cognitive decline will be confirmed using the Mini-Mental State Examination (MMSE). A psychiatrist or psychologist will conduct 30-to-60-min interviews with each participant; these interviews will include free conversation, picture-description task, and story-telling task, all of which will be recorded using a microphone headset. In addition, the severity of disorders will be assessed using clinical rating scales. Data will be collected from each participant at least twice during the study period and up to a maximum of five times at an interval of at least one month. Discussion: This study is unique in its large sample size and the novelty of its method, and has potential for applications in many fields. We have some challenges regarding inter-rater reliability and the linguistic peculiarities of Japanese. As of September 2022, we have collected a total of >1000 records from >400 participants. To the best of our knowledge, this data sample is one of the largest in this field. Clinical Trial Registration: Identifier: UMIN000032141.

Association of work environment with stress and depression among Japanese workers.

BACKGROUND: There is a lack of studies that investigated the effect of a wide range of work environmental factors on stress and depression in Japan. OBJECTIVES: To examine the association of work environment factors with stress and depression among workers in Japan. METHODS: We conducted questionnaire surveys of workers that mainly engage in desk work in Japan. Stress was assessed through the Perceived Stress Scale (PSS), depression through the Patient Health Questionnaire-9 (PHQ-9), and work environment through physical and psychological workplace environment questionnaires. Workers were divided into low and high stress groups based on PSS score (median split), and divided into non-depressed and depressed groups based on their PHQ-9 score (< 5, and ≥5); these groups were then compared with their working environment. In addition, a multiple regression analysis was performed. RESULTS: Responses were obtained from 210 subjects. Multiple regression analysis showed that "Ability to work at one's own pace" and "Ability to apply personal viewpoint to work," etc., had effect on stress, while "Workplace harassment" and "Support from colleagues," etc., had effect on depression. CONCLUSIONS: The results suggest that stress and depression in Japanese workers are related to factors such as job demands, control of work, workplace harassment, and psychological safety.

Evaluation of [18F]PI-2620, a second-generation selective tau tracer, for assessing four-repeat tauopathies.

Tau aggregates represent a key pathologic feature of Alzheimer's disease and other neurodegenerative diseases. Recently, PET probes have been developed for in vivo detection of tau accumulation; however, they are limited because of off-target binding and a reduced ability to detect tau in non-Alzheimer's disease tauopathies. The novel tau PET tracer, [18F]PI-2620, has a high binding affinity and specificity for aggregated tau; therefore, it was hypothesized to have desirable properties for the visualization of tau accumulation in Alzheimer's disease and non-Alzheimer's disease tauopathies. To assess the ability of [18F]PI-2620 to detect regional tau burden in non-Alzheimer's disease tauopathies compared with Alzheimer's disease, patients with progressive supranuclear palsy (n = 3), corticobasal syndrome (n = 2), corticobasal degeneration (n = 1) or Alzheimer's disease (n = 8), and healthy controls (n = 7) were recruited. All participants underwent MRI, amyloid ß assessment and [18F]PI-2620 PET (Image acquisition at 60-90 min post-injection). Cortical and subcortical tau accumulations were assessed by calculating standardized uptake value ratios using [18F]PI-2620 PET. For pathologic validation, tau pathology was assessed using tau immunohistochemistry and compared with [18F]PI-2620 retention in an autopsied case of corticobasal degeneration. In Alzheimer's disease, focal retention of [18F]PI-2620 was evident in the temporal and parietal lobes, precuneus, and cingulate cortex. Standardized uptake value ratio analyses revealed that patients with non-Alzheimer's disease tauopathies had elevated [18F]PI-2620 uptake only in the globus pallidus, as compared to patients with Alzheimer's disease, but not healthy controls. A head-to-head comparison of [18F]PI-2620 and [18F]PM-PBB3, another tau PET probe for possibly visualizing the four-repeat tau pathogenesis in non-Alzheimer's disease, revealed different retention patterns in one subject with progressive supranuclear palsy. Imaging-pathology correlation analysis of the autopsied patient with corticobasal degeneration revealed no significant correlation between [18F]PI-2620 retention in vivo. High [18F]PI-2620 uptake at 60-90 min post-injection in the globus pallidus may be a sign of neurodegeneration in four-repeat tauopathy, but not necessarily practical for diagnosis of non-Alzheimer's disease tauopathies. Collectively, this tracer is a promising tool to detect Alzheimer's disease-tau aggregation. However, late acquisition PET images of [18F]PI-2620 may have limited utility for reliable detection of four-repeat tauopathy because of lack of correlation between post-mortem tau pathology and different retention pattern than the non-Alzheimer's disease-detectable tau radiotracer, [18F]PM-PBB3. A recent study reported that [18F]PI-2620 tracer kinetics curves in four-repeat tauopathies peak earlier (within 30 min) than Alzheimer's disease; therefore, further studies are needed to determine appropriate PET acquisition times that depend on the respective interest regions and diseases.

Subjective well-being and month-long LF/HF ratio among deskworkers.

The importance of workers' well-being has been recognized in recent years. The assessment of well-being has been subjective, and few studies have sought potential biomarkers of well-being to date. This study examined the relationship between well-being and the LF/HF ratio, an index of heart rate variability that reflects sympathetic and parasympathetic nerve activity. Pulse waves were measured using photoplethysmography through a web camera attached to the computer used by each participant. The participants were asked to measure their pulse waves while working for 4 weeks, and well-being was assessed using self-reported measures such as the Satisfaction With Life Scale (SWLS), the Positive and Negative Affect Schedule (PANAS), and the Flourishing Scale (FS). Each of the well-being scores were split into two groups according to the median value, and the LF/HF ratio during work, as well as the number of times an LF/HF ratio threshold was either exceeded or subceeded, were compared between the high and low SWLS, positive emotion, negative emotion, and FS groups. Furthermore, to examine the effects of the LF/HF ratio and demographic characteristics on well-being, a multiple regression analysis was conducted. Data were obtained from 169 participants. The results showed that the low FS group had a higher mean LF/HF ratio during work than the high FS group. No significant differences were seen between the high and low SWLS groups, the high and low positive emotion groups, or the high and low negative emotion groups. The multiple regression analysis showed that the mean LF/HF ratio during work affected the FS and SWLS scores, and the number of times the mean LF/HF ratio exceeded +3 SD had an effect on the positive emotion. No effect of the LF/HF ratio on negative emotions was shown. The LF/HF ratio might be applicable as an objective measure of well-being.

The utility of simple questions to evaluate cognitive impairment.

OBJECTIVES: As the population of patients with cognitive decline grows, physicians and caregivers need brief screening tools. Comprehensive neurocognitive batteries require special training and time for evaluation. We focused on accessibility and compared the diagnostic power of several easy questions. DESIGN: "Attended With" (AW) and "Head-Turning Sign" (HTS) factors and participants' replies to following questions were recorded: "Do you feel that you have more difficulties in your daily life than you used to?", [no consciousness (C-) or consciousness+ (C+)], "Could you tell me about your daily pleasures or pastimes?" [no pleasure (P-) or pleasure + (P+)], "What are notable current/recent news/topics?" [no news (N-) or news+ (N+)]. SETTING: This took place in our Memory Clinic between May 2016 and July 2019. PARTICIPANTS: We enrolled 162 consecutive cases (44 cognitive normal (CN), 55 amnestic mild cognitive impairment (aMCI), and 48 Alzheimer's disease (AD)). MEASUREMENTS: The sensitivity and specificity of each battery were calculated, and on account of those numbers, the population attributable risk percent % (PAR%) of (AW and HTS+), (C- and P-), (C- and N-), (P- and N-) as analysis of combination of questions, respectively, were calculated. RESULTS: AW had high sensitivity, 87.4, 95.8% (CN vs aMCI + AD, CN + aMCI vs AD) but the sensitivity of HTS was only 46.4, 57.7%, and HTS showed high specificity, 100.0, 71.8%. C- had high sensitivity, 80.6, 87.5%, whereas P- and N- had high specificity, both 83.9% in CN vs aMCI + AD, 88.1% and 75.9% in CN + aMCI vs AD, respectively. In combination analysis, the PAR% of (C- and N-) were as high as (AW and HTS+). CONCLUSIONS: The combination of (C- and N-) is as powerful as (AW and HTS+) in screening AD. Our findings provide novel insights for screening utility of brief questions "Consciousness of Impairment" and "Recent News."