Predicting changes in brain metabolism and progression from mild cognitive impairment to dementia using multitask Deep Learning models and explainable AI.

BACKGROUND: The prediction of Alzheimer's disease (AD) progression from its early stages is a research priority. In this context, the use of Artificial Intelligence (AI) in AD has experienced a notable surge in recent years. However, existing investigations predominantly concentrate on distinguishing clinical phenotypes through cross-sectional approaches. This study aims to investigate the potential of modeling additional dimensions of the disease, such as variations in brain metabolism assessed via [18F]-fluorodeoxyglucose positron emission tomography (FDG-PET), and utilize this information to identify patients with mild cognitive impairment (MCI) who will progress to dementia (pMCI). METHODS: We analyzed data from 1,617 participants from the Alzheimer's Disease Neuroimaging Initiative (ADNI) who had undergone at least one FDG-PET scan. We identified the brain regions with the most significant hypometabolism in AD and used Deep Learning (DL) models to predict future changes in brain metabolism. The best-performing model was then adapted under a multi-task learning framework to identify pMCI individuals. Finally, this model underwent further analysis using eXplainable AI (XAI) techniques. RESULTS: Our results confirm a strong association between hypometabolism, disease progression, and cognitive decline. Furthermore, we demonstrated that integrating data on changes in brain metabolism during training enhanced the models' ability to detect pMCI individuals (sensitivity=88.4%, specificity=86.9%). Lastly, the application of XAI techniques enabled us to delve into the brain regions with the most significant impact on model predictions, highlighting the importance of the hippocampus, cingulate cortex, and some subcortical structures. CONCLUSION: This study introduces a novel dimension to predictive modeling in AD, emphasizing the importance of projecting variations in brain metabolism under a multi-task learning paradigm.

[Diabetes control and visits to the family doctor during the COVID-19 pandemic: a multicenter study in primary care]. / Control de diabetes y consultas al médico de familia durante la pandemia de COVID-19: estudio multicéntrico en atención primaria.

OBJECTIVES: To determine whether in patients with type 2 diabetes (DM2) the changes in their relationship with family doctors during the COVID-19 pandemic, in-person (iPC) and telematic (TC) consultations, were associated with control of their disease. DESIGN: Multicentric study of retrospective follow-up. SETTING: Seven health centers in Tenerife, Spain. PARTICIPANTS: 3543 patients with DM2. MAIN MEASUREMENTS: Sex, age, iPC, TC and DM2 control using glycosylated hemoglobin (A1c) during the period 2019-2021. Logistic regression models were fitted with DM2 control as an effect, and with the other measurements as independent variables. RESULTS: 50% were women. 38% were less than 65 years old. A1c was measured in 84% of patients in 2019, 68% in 2020, and 77% in 2021. 58.4% had good control in 2019, 46.1% in 2020, and 50.3% in 2021. Median iPC were 7 in 2019, 4 in 2020 and 5 in 2021 (p<0.001). The OR(95%CI) of good control in 2019 were 1.04(1.04-1.05) per year of age and 1.03(1.01-1.04) for each iPC; In 2020 they were 1.04 (1.03-1.05) per year of age, 1.05 (1.04-1.07) for each iPC and 1.04 (1.02-1.07) for each TC; in 2021 they were 1.04 (1.04-1.05) per year of age, 1.05 (1.03-1.06) for each iPC and 1.02 (1.00-1.04) for each TC. CONCLUSIONS: The control of patients with DM2 during the period 2019-2021 had a direct relationship with the change in the frequency of consultations at the health center, with differences depending on the type of consultation and the age of the patient.

Transcranial Doppler 6 h after Successful Reperfusion as a Predictor of Infarct Volume.

OBJECTIVES: The aim of the study is to analyze the hemodynamic changes in the middle cerebral artery (MCA) after endovascular revascularization in acute ischemic stroke (AIS) due to large vessel occlusion and its association with the infarct volume size in the control head CT. MATERIALS AND METHODS: Prospective study of patients with AIS due to internal carotid artery terminus or M1 segment of the MCA occlusion, who underwent endovascular treatment with a final TICI 2b-3 score, without concomitant stenosis ≥50% in both cervical carotid arteries. Transcranial Doppler ultrasound (TCD) of both MCAs was carried out at 6 h after the endovascular procedure. Mean flow velocities (MFV) after arterial reperfusion and its association with the infarct volume size in 24-36 h control head CT were determined. RESULTS: 91 patients (51 women) were included with a median age of 78 years and National institute of Health Stroke Scale of 18. The MCA was occluded in 76.92%, and intravenous thrombolysis was administered in 40.7%. The incidence of symptomatic intracranial hemorrhage was 5.5%. At three months, mortality was 19.8% and a 52.7% of patients achieved functional independence (modified Rankin Scale 0-2). After a multivariable logistic regression analysis, an increase in the MFV greater than 50% at 6 h in the treated MCA compared to contralateral MCA, was an independent predictor of large infarct volume in the control head CT with an OR 9.615 (95%CI: 1.908-47.620), p=0.006 CONCLUSIONS: Increased MFV assessed by TCD examination following endovascular recanalization is independently associated with larger infarct volume.

ACE2, TMPRSS2, and Furin variants and SARS-CoV-2 infection in Madrid, Spain.

It has been suggested that some individuals may present genetic susceptibility to SARS-CoV-2 infection, with particular research interest in variants of the ACE2 and TMPRSS2 genes, involved in viral penetration into cells, in different populations and geographic regions, although insufficient information is currently available. This study addresses the apparently reasonable hypothesis that variants of these genes may modulate viral infectivity, making some individuals more vulnerable than others. Through whole-exome sequencing, the frequency of exonic variants of the ACE2, TMPRSS2, and Furin genes was analyzed in relation to presence or absence of SARS-CoV-2 infection in a familial multiple sclerosis cohort including 120 individuals from Madrid. The ACE2 gene showed a low level of polymorphism, and none variant was significantly associated with SARS-CoV-2 infection. These variants have previously been detected in Italy. While TMPRSS2 is highly polymorphic, the variants found do not coincide with those described in other studies, with the exception of rs75603675, which may be associated with SARS-CoV-2 infection. The synonymous variants rs61735792 and rs61735794 showed a significant association with infection. Despite the limited number of patients with SARS-CoV-2 infection, some variants, especially in TMPRSS2, may be associated with COVID-19.

Diagnosis of Alzheimer's disease and behavioural variant frontotemporal dementia with machine learning-aided neuropsychological assessment using feature engineering and genetic algorithms.

BACKGROUND: Neuropsychological assessment is considered a valid tool in the diagnosis of neurodegenerative disorders. However, there is an important overlap in cognitive profiles between Alzheimer's disease (AD) and behavioural variant frontotemporal dementia (bvFTD), and the usefulness in diagnosis is uncertain. We aimed to develop machine learning-based models for the diagnosis using cognitive tests. METHODS: Three hundred and twenty-nine participants (170 AD, 72 bvFTD, 87 healthy control [HC]) were enrolled. Evolutionary algorithms, inspired by the process of natural selection, were applied for both mono-objective and multi-objective classification and feature selection. Classical algorithms (NativeBayes, Support Vector Machines, among others) were also used, and a meta-model strategy. RESULTS: Accuracies for the diagnosis of AD, bvFTD and the differential diagnosis between them were higher than 84%. Algorithms were able to significantly reduce the number of tests and scores needed. Free and Cued Selective Reminding Test, verbal fluency and Addenbrooke's Cognitive Examination were amongst the most meaningful tests. CONCLUSIONS: Our study found high levels of accuracy for diagnosis using exclusively neuropsychological tests, which supports the usefulness of cognitive assessment in diagnosis. Machine learning may have a role in improving the interpretation and test selection.

Regeneration and reuse of magnetic particles for contaminant degradation in water.

Fenton reaction is an oxidation process of interest in wastewater treatment because of its ability to degrade organic compounds. Iron-based magnetic particles can be a very useful catalyst when using heterogeneous Fenton process. The major problem of this heterogeneous process is the saturation of the Fe 3+ on the surface, which limits the process. In this study, the possibility of using magnetite particles as a substrate is presented, increasing its degradation efficiency by Fenton reaction through a regeneration process that achieves the electronic reduction of its surface using reducing agents. The results indicate that the regeneration process is quite effective, increasing the efficiency of the degradation of Methylene Blue up to 99%. The concentration of magnetite is the most influential factor in the efficiency of the reaction, while the regeneration time and the concentration of reducing agent do not significantly affect the results considering the range used. The presence of mechanical stirring may adversely affect the reaction in the long term. Increasing the oxidant agent concentration reduces the initial speed of the reaction but not the long-term efficiency. The use of hydrazine in this process allows the successive reuse of these particles maintaining a high percentage of elimination of methylene blue, above 70% even after 10 uses, compared to an elimination below 20% for particles not regenerated after the second use and for particles regenerated with ascorbic acid after the eighth use.

Magnetic water treatment in a wastewater treatment plant: Part I - sorption and magnetic particles.

The application of magnetic sorption to treat wastewaters is nowadays seen as a potential industrial method. In this work we apply magnetite particles to remediate real wastewater samples, with several contaminants competing for the same active sorption center at the same time. We also apply our studies at three different sampling points of a Wastewater Treatment Plant. In general terms, magnetite particles have shown a very good behaviour concerning the reduction of detergents and COD, while phosphates and total nitrogen, and the majority of heavy metals are high to moderately removed. The influence of the type of wastewater (i.e., sampling point) has also shown to be important especially for high concentration of contaminants.

Magnetic water treatment in a wastewater treatment plant: Part II - Processing waters and kinetic study.

Magnetic sorption process is applied to real wastewater effluents from a Wastewater Treatment Plant. The complex media sorption is done by using different types of magnetic particles (resin and polymeric covered magnetite) giving good results for removing detergents, phosphates and COD and moderate results for the sorption of nitrogen and several heavy metals. Important kinetic parameters were obtained by data fitting for the pseudo first and second order, and for simplified Elovich models. Regeneration and reuse of the magnetic particles using a chemical-free method was also tested, as well as the effect of the concentration of the particles in the removal efficiency (which proved to be relevant).

Lacrimal Nerve Blocks for Three New Cases of Lacrimal Neuralgia.

OBJECTIVE: Our aim was to report three new cases of lacrimal neuralgia and their response to superficial nerve blocks. BACKGROUND: Lacrimal neuralgia has been recently described as a pain in the territory supplied by the lacrimal nerve, at the lateral upper eyelid and/or the adjacent area of the temple. The pain is typically accompanied by tenderness on palpation of the lacrimal nerve at the superoexternal angle of the orbit. METHODS: Between January 2015 and June 2016, we prospective identified three cases of lacrimal neuralgia among the patients attending the Headache Unit in a tertiary hospital. Anesthetic blocks were performed in all cases by inserting a 30-G needle on the emergence of the nerve and injecting 0.5 cc of bupivacaine 0.5% subcutaneously. RESULTS: Three women aged 44, 49, and 51 presented with pain in the territory supplied by the lacrimal nerve. Two of them localized their pain in a small area of the right temple, while the remaining patient had pain in the right upper lateral eyelid and a small area of the lower lateral eyelid. The pain was continuous in two patients and episodic with attacks lasting 48 hours in one patient. All patients had tenderness on palpation of the lacrimal nerve. Anesthetic blocks confirmed the diagnosis of lacrimal neuralgia and provided the patients with long-lasting pain relief. CONCLUSIONS: Lacrimal neuralgia should be considered among the neuralgic causes of orbital and periorbital pain. Superficial nerve blocks may assist clinicians in the diagnosis and may also be a therapeutic option.

Comparison of QTc and Troponin Levels in ST Elevation MIs Compared with Non-ST Elevation MIs.

OBJECTIVES: ST elevation myocardial infarctions (STEMIs) and non-ST elevation myocardial infarctions (NSTEMIs) have differences that can be important to differentiate. Our primary hypothesis was that corrected QT (QTc) duration and troponin I levels were higher in STEMIs compared with NSTEMIs. The objective of our study was to compare STEMIs with NSTEMIs for QTc duration and troponin levels. METHODS: This was a retrospective case-control study of all STEMIs and a random sample of NSTEMIs during a 1-year period. STEMIs were retrieved by searching our electrocardiogram database for all of the cardiology-diagnosed STEMIs. NSTEMIs were found by selecting a randomized sample of all of the patients with a final discharge diagnosis of NSTEMI. Records and electrocardiograms were reviewed for initial troponin I levels and QTc duration. Data extractors were educated formally and a 5% sample was reevaluated by the other extractor as a reliability measure. Data analysis included χ2 tests and parametric or nonparametric analysis, where appropriate. A logistic regression model was created with variables selected a priori for predictors of STEMIs compared with NSTEMIs. RESULTS: A total of 92 STEMIs and 111 NSTEMIs were evaluated, and interrater reliability showed 90% agreement. Patients with NSTEMIs had significantly longer QTc. Troponin I did not differ on univariate analysis. In a logistic model, Hispanics were more likely than whites to have a STEMI (adjusted odds ratio [AOR] 2.2, 95% confidence interval [CI] 1.09-4.5). An increase in troponin I of 1 was associated with a 7% increase in the AOR of a STEMI (AOR 1.7, 95% CI 1.03-1.12) and an increase in QTc by 10 was associated with a 13% decrease in the AOR of a STEMI (AOR 0.87, 95% CI 0.78-0.93). CONCLUSIONS: Patients with NSTEMIs had longer QTc intervals and lower troponin I levels than those with STEMIs.

The meaning of elevated troponin I levels: not always acute coronary syndromes.

BACKGROUND: Troponin elevation can be caused by etiologies other than acute coronary syndromes (ACS). Our hypothesis was that elevated troponins occur more frequently in non-ACS cases but that ACS cases (type 1 ST-elevation myocardial infarction [STEMI] and type 1 non-STEMI [NSTEMI]) have significantly higher troponin elevations. METHODS: This was a cross-sectional cohort analysis of a random subset of all patients with elevated troponins (defined as ≥0.06 ng/mL) over a 1-year period from July 2013 to June 2014. The first positive troponin I and the peak were used in this study. All included patients had medical record reviews looking for whether our cardiologists or hospitalists attributed the elevated troponin to an ACS (NSTEMI or STEMI) or non-ACS cause. Non-ACS causes were categorized as infection, cancer, renal diseases, cardiovascular disease, pulmonary disease, trauma, cardiac arrest, neurologic disease, hypertension, or other. Data were extracted by 2 investigators on the cause of the elevated troponin. Three sessions to educate data extractors were arranged and methods of data extraction discussed, then a 5% sample was reevaluated by the other extractor to determine interrater agreement measures. Parametric data were evaluated with t test and analysis of variance. Dichotomous variables were compared using χ(2) test. Troponin data were evaluated using nonparametric Kruskal-Wallis or Mann-Whitney U. A logistic regression model was created with variables selected a priori to evaluate the predictive ability of these variables in differentiating ACS vs non-ACS causes of elevated troponin. RESULTS: We evaluated 458 randomly selected patients from 1317 unique cases of all patients with initial elevated troponins at least 0.06 mg/mL during the study period. There was 84% interrater agreement in the 5% sampling. Seventy-nine percent had a non-ACS cause of elevated troponin, and the average initial positive troponin I level was significantly lower in the non-ACS cases (0.14; 95% confidence interval [CI], 0.08-0.37) than those with documented STEMI (10.2; 95% CI, 0.75-20.1) or NSTEMIs (0.4; 95% CI, 0.13-1.7). In the non-ACS group, the median initial troponin was 0.14 ng/mL (0.08-0.37 ng/mL). Peak troponin levels were highest in STEMI, next NSTEMI, and lowest in non ACS causes. The most frequent subgroups in the non-ACS group were non-ACS cardiovascular, infectious, renal, or hypertensive causes. In a linear regression model adjusting for age and sex, higher troponin levels had higher odds of being related to ACS causes (adjusted odds ratio, 1.4; 95% CI, 1.2-1.6) than non-ACS causes. CONCLUSION: The etiology for most initial elevated troponin I levels in a randomly selected population is the result of non-ACS causes. As initial + troponin levels increased, they were more likely associated with ACS causes than with non-ACS causes. Average initial + and peak troponin values were highest in STEMIs, next highest in NSTEMIs, and lowest overall in non-ACS causes.

Nisin Inhibition of Gram-Negative Bacteria.

Aims: This study investigates the activity of the broad-spectrum bacteriocin nisin against a large panel of Gram-negative bacterial isolates, including relevant plant, animal, and human pathogens. The aim is to generate supportive evidence towards the use/inclusion of bacteriocin-based therapeutics and open avenues for their continued development. Methods and Results: Nisin inhibitory activity was screened against a panel of 575 strains of Gram-negative bacteria, encompassing 17 genera. Nisin inhibition was observed in 309 out of 575 strains, challenging the prevailing belief that nisin lacks effectiveness against Gram-negative bacteria. The genera Acinetobacter, Helicobacter, Erwinia, and Xanthomonas exhibited particularly high nisin sensitivity. Conclusions: The findings of this study highlight the promising potential of nisin as a therapeutic agent for several key Gram-negative plant, animal, and human pathogens. These results challenge the prevailing notion that nisin is less effective or ineffective against Gram-negative pathogens when compared to Gram-positive pathogens and support future pursuits of nisin as a complementary therapy to existing antibiotics. Significance and Impact of Study: This research supports further exploration of nisin as a promising therapeutic agent for numerous human, animal, and plant health applications, offering a complementary tool for infection control in the face of multidrug-resistant bacteria.

ORAI1 inhibition as an efficient preclinical therapy for tubular aggregate myopathy and Stormorken syndrome.

Tubular aggregate myopathy (TAM) and Stormorken syndrome (STRMK) are clinically overlapping disorders characterized by childhood-onset muscle weakness and a variable occurrence of multisystemic signs, including short stature, thrombocytopenia, and hyposplenism. TAM/STRMK is caused by gain-of-function mutations in the Ca2+ sensor STIM1 or the Ca2+ channel ORAI1, both of which regulate Ca2+ homeostasis through the ubiquitous store-operated Ca2+ entry (SOCE) mechanism. Functional experiments in cells have demonstrated that the TAM/STRMK mutations induce SOCE overactivation, resulting in excessive influx of extracellular Ca2+. There is currently no treatment for TAM/STRMK, but SOCE is amenable to manipulation. Here, we crossed Stim1R304W/+ mice harboring the most common TAM/STRMK mutation with Orai1R93W/+ mice carrying an ORAI1 mutation partially obstructing Ca2+ influx. Compared with Stim1R304W/+ littermates, Stim1R304W/+Orai1R93W/+ offspring showed a normalization of bone architecture, spleen histology, and muscle morphology; an increase of thrombocytes; and improved muscle contraction and relaxation kinetics. Accordingly, comparative RNA-Seq detected more than 1,200 dysregulated genes in Stim1R304W/+ muscle and revealed a major restoration of gene expression in Stim1R304W/+Orai1R93W/+ mice. Altogether, we provide physiological, morphological, functional, and molecular data highlighting the therapeutic potential of ORAI1 inhibition to rescue the multisystemic TAM/STRMK signs, and we identified myostatin as a promising biomarker for TAM/STRMK in humans and mice.

Genetic-based patient stratification in Alzheimer's disease.

Alzheimer's disease (AD) shows a high pathological and symptomatological heterogeneity. To study this heterogeneity, we have developed a patient stratification technique based on one of the most significant risk factors for the development of AD: genetics. We addressed this challenge by including network biology concepts, mapping genetic variants data into a brain-specific protein-protein interaction (PPI) network, and obtaining individualized PPI scores that we then used as input for a clustering technique. We then phenotyped each obtained cluster regarding genetics, sociodemographics, biomarkers, fluorodeoxyglucose-positron emission tomography (FDG-PET) imaging, and neurocognitive assessments. We found three clusters defined mainly by genetic variants found in MAPT, APP, and APOE, considering known variants associated with AD and other neurodegenerative disease genetic architectures. Profiling of these clusters revealed minimal variation in AD symptoms and pathology, suggesting different biological mechanisms may activate the neurodegeneration and pathobiological patterns behind AD and result in similar clinical and pathological presentations, even a shared disease diagnosis. Lastly, our research highlighted MAPT, APP, and APOE as key genes where these genetic distinctions manifest, suggesting them as potential targets for personalized drug development strategies to address each AD subgroup individually.

A data-driven approach to complement the A/T/(N) classification system using CSF biomarkers.

AIMS: The AT(N) classification system not only improved the biological characterization of Alzheimer's disease (AD) but also raised challenges for its clinical application. Unbiased, data-driven techniques such as clustering may help optimize it, rendering informative categories on biomarkers' values. METHODS: We compared the diagnostic and prognostic abilities of CSF biomarkers clustering results against their AT(N) classification. We studied clinical (patients from our center) and research (Alzheimer's Disease Neuroimaging Initiative) cohorts. The studied CSF biomarkers included Aß(1-42), Aß(1-42)/Aß(1-40) ratio, tTau, and pTau. RESULTS: The optimal solution yielded three clusters in both cohorts, significantly different in diagnosis, AT(N) classification, values distribution, and survival. We defined these three CSF groups as (i) non-defined or unrelated to AD, (ii) early stages and/or more delayed risk of conversion to dementia, and (iii) more severe cognitive impairment subjects with faster progression to dementia. CONCLUSION: We propose this data-driven three-group classification as a meaningful and straightforward approach to evaluating the risk of conversion to dementia, complementary to the AT(N) system classification.

Gender, life events, and mental well-being in emerging adulthood.

BACKGROUND: Emerging adulthood is a critical period of life involving many life transitions that may generate stress and compromise health and mental well-being. AIMS: To know the most frequent life events of women and men in emerging adulthood, analyzing also the relevance that such stressors have on their psychological well-being and life satisfaction. A second aim is to determine the relevance of age, educational level, most frequent life events, coping styles, and perceived social support as risk and protective factors for well-being. METHOD: The sample consisted of 2,000 individuals from the general Spanish population (55% women), aged between 18 and 29, who were assessed using five questionnaires and scales measuring life events, coping styles, life satisfaction, psychological well-being, and social support. RESULTS: The findings showed that 90% of the sample had experienced one or more life events during the previous year and that a higher number of life events experienced over the past year was associated with lower mental well-being. Multiple regression analyses made clear that, although some events experienced in the previous year (namely, family conflicts and change in the relationship with parents) were associated with lower women's and men's well-being, the most important determinants of well-being in either gender were coping styles; however, some predictors of women's well-being proved different from those of men. In the case of women, family and intimate partner conflicts predicted lower life satisfaction and psychological well-being was lower in the case of family conflicts. As for men, work or academic life events predicted lower life satisfaction. CONCLUSION: These research findings are relevant for the design of programs and strategies to improve mental well-being in emerging adulthood.

Dating Violence and Mental Health in Emerging Adulthood.

Dating violence constitutes a serious social and health problem. This study aims to increase knowledge on dating violence in emerging adulthood by analysing the relevance of gender and of having or not having a current partner in the victimization and perpetration of such violence. It also analyses the association between dating violence and mental health, as well as the relevance of traditional gender role attitudes and the internalization of feminine/expressive and masculine/instrumental traits in the victimization and perpetration of such types of violence. The participants were 930 Spanish emerging adults who were assessed by six self-report questionnaires and scales. Men reported more psychological and physical violence victimization and physical violence perpetration than women, and women and men without a current partner reported more psychological and sexual violence than women and men with a current partner. Dating violence victimization was associated with more mental symptomatology, less life satisfaction, and lower self-esteem in men with a current partner and in women without a current partner. The main predictor of dating violence victimization was dating violence perpetration, and the main predictor of dating violence perpetration was victimization by such violence. More traditional gender role attitudes also predicted greater victimization and perpetration of dating violence, except among women without a current partner.

Multi-Disease Validation of the RUDAS for Cognitive Screening in Alzheimer's Disease, Parkinson's Disease, and Multiple Sclerosis.

BACKGROUND: The Rowland Universal Dementia Assessment Scale (RUDAS) is a cognitive test with favorable diagnostic properties for detecting dementia and a low influence of education and cultural biases. OBJECTIVE: We aimed to validate the RUDAS in people with Alzheimer's disease (AD), Parkinson's disease (PD), and multiple sclerosis (MS). METHODS: We enrolled one hundred and fifty participants (60 with AD, 30 with PD, 60 with MS, and 120 healthy controls (HC)). All clinical groups completed a comprehensive neuropsychological battery, RUDAS, and standard cognitive tests of each disorder: MMSE, SCOPA-COG, and Symbol Digit Modalities Test. Intergroup comparisons between clinical groups and HC and ROC curves were estimated. Random Forest algorithms were trained and validated to detect cognitive impairment using RUDAS and rank the most relevant scores. RESULTS: The RUDAS scores were lower in patients with AD, and patients with PD and MS showed cognitive impairment compared to healthy controls. Effect sizes were generally large. The total score was the most discriminative, followed by the memory score. Correlations with standardized neuropsychological tests were moderate to high. Random Forest algorithms obtained accuracies over 80-90% using the RUDAS for diagnosing AD and cognitive impairment associated with PD and MS. CONCLUSION: Our results suggest the RUDAS is a valid test candidate for multi-disease cognitive screening tool in AD, PD, and MS.

Partial Thrombosis of Corpus Cavernosum.

Thrombosis of the corpus cavernosum is a rare disease of unknown cause that usually affects young men. We present the case of a 25-year-old man with an unilateral, painful perineal mass and ultrasound scan compatible with this entity. The magnetic resonance and tomography computarised scan images corroborate the diagnosis. Anticoagulants were prescribed which solved the clinical picture.

On the limits of graph neural networks for the early diagnosis of Alzheimer's disease.

Alzheimer's disease (AD) is a neurodegenerative disease whose molecular mechanisms are activated several years before cognitive symptoms appear. Genotype-based prediction of the phenotype is thus a key challenge for the early diagnosis of AD. Machine learning techniques that have been proposed to address this challenge do not consider known biological interactions between the genes used as input features, thus neglecting important information about the disease mechanisms at play. To mitigate this, we first extracted AD subnetworks from several protein-protein interaction (PPI) databases and labeled these with genotype information (number of missense variants) to make them patient-specific. Next, we trained Graph Neural Networks (GNNs) on the patient-specific networks for phenotype prediction. We tested different PPI databases and compared the performance of the GNN models to baseline models using classical machine learning techniques, as well as randomized networks and input datasets. The overall results showed that GNNs could not outperform a baseline predictor only using the APOE gene, suggesting that missense variants are not sufficient to explain disease risk beyond the APOE status. Nevertheless, our results show that GNNs outperformed other machine learning techniques and that protein-protein interactions lead to superior results compared to randomized networks. These findings highlight that gene interactions are a valuable source of information in predicting disease status.