Personalized Diagnosis Research Articles

A personalized federated meta-learning method for intelligent and privacy-preserving fault diagnosis

Intelligent fault diagnosis methods have made significant progress during the past decade, often following a centralized training paradigm that gathers data from multiple industrial users, which results in high acquisition costs and raise data privacy concerns. Federated Learning (FL) provides a promising solution to intrinsically protect data privacy by jointly training a shared global model without sharing individual data, but the varying data distributions among clients pose challenges for diagnostic performance. To address these problems, a novel FL framework called federated meta-learning based on fine-grained classifier reconstruction (FedFGCR) is presented in this paper. First, an adaptive interpolation strategy based on meta-learning is designed to compute the optimal combination of global and local feature extractors as the initialization of the local training. Additionally, we introduce a new regularization term that allows the local model to leverage global data, enhancing model generalization by incorporating global semantic feature information. Furthermore, a fine-grained classifier reconstruction method is proposed to improve the personalized diagnosis ability for local samples. Comparative experiments with mainstream FL-based fault diagnosis methods have been conducted, demonstrating FedFGCR’s superiority with accuracy gains ranging from 0.96% to 5.41% on the CWRU dataset and from 5.01% to 9.95% on the KAIST dataset.

Individualized functional connectome abnormalities obtained using two normative model unveil neurophysiological subtypes of obsessive compulsive disorder

The high heterogeneity observed among patients with obsessive-compulsive disorder (OCD) underscores the need to identify neurophysiological OCD subtypes to facilitate personalized diagnosis and treatment. In this study, our aim was to identify potential OCD subtypes based on individualized functional connectome abnormalities. We recruited a total of 99 patients with OCD and 104 healthy controls (HCs) matched for demographic characteristics. Individualized functional connectome abnormalities were obtained using normative models of functional connectivity strength (FCS) and used as features to unveil OCD subtypes. Sensitivity analyses were conducted to assess the reproducibility and robustness of the clustering outcomes. Patients exhibited significant intersubject heterogeneity in individualized functional connectome abnormalities. Two subtypes with distinct patterns of FCS abnormalities relative to HCs were identified. Subtype 1 patients primarily exhibited significantly decreased FCS in regions including the frontal gyrus, insula, hippocampus, and precentral/postcentral gyrus, whereas subtype 2 patients demonstrated increased FCS in widespread brain regions. When all patients were combined, no significant differences were observed. Additionally, the identified subtypes showed significant differences in age of onset. Furthermore, sensitivity analyses confirmed the reproducibility of our subtyping results. In conclusion, the identified OCD subtypes shed new light on the taxonomy and neurophysiological heterogeneity of OCD.

Label-guided low-rank Approximation for functional brain network learning in identifying subcortical vascular cognitive impairment

Background and ObjectiveFunctional brain network (FBN) has played a pivotal role in unraveling the inherent mechanism of cognition/behavior and investigating neuropsychological disorders. Particularly, growing evidence suggests that FBNs have shown great potential in classifying brain disorders such as subcortical vascular cognitive impairment (SVCI). However, learning a high-quality FBN is still challenging since there is no ground truth. Most traditional methods tend to focus on learning FBNs independently of the downstream task. This practice neglects the valuable label information crucial for accurate classification. Besides, the methodology of constructing brain networks in an isolated, individualistic manner sidesteps the common information among individuals. MethodsTo address these issues, we developed a new FBN joint learning strategy named Label-Guided Low-rank Approximation (LGLA). LGLA integrates label information from training subjects and uses unlabeled testing subjects for auxiliary training, enhancing the discriminative power of FBN features tailored to testing subjects. By enforcing class-specific similarities and differences, and applying a low-rank constraint to capture shared information, LGLA presents an optimization framework that can be effectively solved. ResultsExperimental results demonstrate the effectiveness of our proposed FBN learning method for identifying SVCI. The proposed method achieved an accuracy of 75.70% and an area under the receiver operating characteristic curve (AUC) of 0.8233 in the task of identifying patients with amnestic mild cognitive impairment (aMCI). For identifying patients with non-amnestic mild cognitive impairment (naMCI), it achieved an accuracy of 63.64% and an AUC of 0.7857, while the task of identifying patients with no cognitive impairment (NCI), it achieved an accuracy of 63.64% and an AUC of 0.7917. Additionally, we further explored the brain network features and discovered potential biomarkers for personalized diagnosis. ConclusionThe LGLA method enhances FBN construction and classification accuracy for SVCI diagnosis by effectively integrating the label information and shared structure among subjects.

Impact of Mental Health Court Diversion on Reoffending: A Direct Comparison of Diverted and Undiverted Groups

Mental health court diversionary programs are a key strategy to address over-representation of individuals with a mental illness in the criminal justice system. The study examined 2,476 individuals identified as eligible for mental health court diversion by the Statewide Community and Court Liaison Service (SCCLS) in New South Wales (NSW), Australia and compared reoffending of those granted and not granted court diversion. Individuals not diverted had 43% higher reoffending rate than those granted diversion. Predictors of non-diversion and reoffending included younger age, being male, Aboriginal and/or Torres Strait Islander background, primary personality and/or substance use diagnosis and non-violent offence charges.

Educational and occupational aspirations of adolescent and young adult cancer survivors: a qualitative analysis.

Adolescent and young adult cancer survivors (AYACS) are patients diagnosed with cancer between 15 and 39years of age. AYACS are often derailed from planned educational and occupational endeavors due to disruption from cancer treatment and its consequences. The study objective was to examine how a personal cancer diagnosis impacted AYACS' experiences related to these endeavors. Semi-structured interviews were conducted as part of a larger study assessing psychosocial challenges among a younger AYACS subset aged 15-25years old at the time of cancer diagnosis. Interviews were coded based on responses and were used to develop themes related to educational and occupational endeavors. Data were collected from 35 participants. Five themes emerged: (1) Pauses in educational attainment had a detrimental effect on educational goals for some participants, but further solidified and sculpted educational plans for others; (2) Although participants experienced challenges accomplishing educational goals, supportive school environments helped surmount these challenges; (3) Participants reflected on rethinking career aspirations, though some desired to pursue the same occupation planned before cancer diagnosis; (4) Participants experienced challenges, including physical and cognitive limitations, upon returning to work; and (5) Participants valued autonomy and normalcy through work and appreciated supportive and flexible work environments. AYACS prioritize professional achievement, yet encounter challenges in achieving professional goals. Our findings create a foundation for developing and testing prospective interventions to promote continuance of school and work during cancer treatment when feasible, and proactive reintegration strategies for those who paused professional goals due to cancer treatment.

A Review on Autoimmune Diseases: Recent Advances and Future Perspectives

Autoimmune diseases represent a diverse group of disorders characterized by immune-mediated attacks on the body’s own tissues and organs. This review provides a comprehensive overview of autoimmune diseases, covering their definition, classification, epidemiology, etiology, pathogenesis, clinical manifestations, current treatments, and future research directions. Autoimmune diseases can be categorized into organ-specific (e.g., type 1 diabetes, Hashimoto's thyroiditis) and systemic (e.g., systemic lupus erythematosus, rheumatoid arthritis) conditions, each with distinct clinical presentations and underlying mechanisms. Genetic predisposition, environmental triggers, and immunological dysregulation play critical roles in disease development. Common symptoms include fatigue, joint pain, skin manifestations, and organ-specific dysfunction, contributing to significant morbidity and impaired quality of life. Current treatment strategies encompass immunosuppressive therapies, disease-modifying drugs, and emerging biologic agents targeting specific immune pathways. Advances in genomics, immunology, and precision medicine offer promising avenues for personalized diagnosis and treatment optimization. Future research directions include further elucidating disease heterogeneity, identifying novel biomarkers, and developing targeted immunotherapies to achieve long-term remission and improve patient outcomes. Understanding the complex interplay of genetic, environmental, and immunological factors is crucial for advancing therapeutic approaches and mitigating the global burden of autoimmune diseases. Enhanced collaboration across disciplines and continued investment in research are essential to translate these insights into clinical practice and benefit patients worldwide.

IgE-Mediated Cannabis Allergy and Cross-Reactivity Syndromes: A Roadmap for Correct Diagnosis and Management.

With increased access and decriminalization of cannabis use, cases of IgE-dependent cannabis allergy (CA) and cross-reactivity syndromes have been increasingly reported. However, the exact prevalence of cannabis allergy and associated cross-reactive food syndromes (CAFS) remains unknown and is likely to be underestimated due to a lack of awareness and insufficient knowledge of the subject among health care professionals. Therefore, this practical roadmap aims to familiarize the reader with the early recognition and correct management of IgE-dependent cannabis-related allergies. In order to understand the mechanisms underlying these cross-reactivity syndromes and to enable personalized diagnosis and management, special attention is given to the molecular diagnosis of cannabis-related allergies. The predominant signs and symptoms of CA are rhinoconjunctivitis and contact urticaria/angioedema. However, CA can also present as a life-threatening condition. In addition, many patients with CA also have distinct cross-reactivity syndromes, mainly involving fruits, vegetables, nuts and cereals. At present, five allergenic components of Cannabis sativa (Can s); Can s 2 (profilin), Can s 3 (a non-specific lipid protein), Can s 4 (oxygen-evolving enhancer protein 2 oxygen), Can s 5 (the Bet v 1 homologue) and Can s 7 (thaumatin-like protein) have been characterized and indexed in the WHO International Union of Immunological Sciences (IUIS) allergen database. However, neither of them is currently readily available for diagnosis, which generally starts by testing crude extracts of native allergens. The road to a clear understanding of CA and the associated cross-reactive food syndromes (CAFS) is still long and winding, but well worth further exploration.

Synergistic Strategies of Biomolecular Transport Technologies in Transdermal Healthcare Systems.

Transdermal healthcare systems have gained significant attention for their painless and convenient drug administration, as well as their ability to detect biomarkers promptly. However, the skin barrier limits the candidates of biomolecules that can be transported, and reliance on simple diffusion poses a bottleneck for personalized diagnosis and treatment. Consequently, recent advancements in transdermal transport technologies have evolved toward active methods based on external energy sources. Multiple combinations of these technologies have also shown promise for increasing therapeutic effectiveness and diagnostic accuracy as delivery efficiency is maximized. Furthermore, wearable healthcare platforms are being developed in diverse aspects for patient convenience, safety, and on-demand treatment. Herein, a comprehensive overview of active transdermal delivery technologies is provided, highlighting the combination-based diagnostics, therapeutics, and theragnostics, along with the latest trends in platform advancements. This offers insights into the potential applications of next-generation wearable transdermal medical devices for personalized autonomous healthcare.

Association of psychiatric and substance use disorders with cannabis use and cannabis use disorder during early pregnancy in northern California.

To estimate the strength of association between psychiatric disorders and substance use disorders (SUD), and cannabis use and cannabis use disorder (CUD)during early pregnancy. Observational study. Kaiser Permanente Northern California, USA. 299 496 pregnancies from 227 555 individuals screened for cannabis use by self-report and a urine toxicology test at entrance to prenatal care in Kaiser Permanente Northern California during January 2011-December 2021 (excepting year 2020). The sample was 62.5% non-White, with a mean (standard deviation) age of 31.1 (5.5) years; 6.8% used cannabis; 0.2% had a CUD. Exposure variables included electronic health record-based psychiatric diagnoses of attention deficit hyperactivity, anxiety, bipolar, depressive, personality, posttraumatic stress and psychotic disorders; and alcohol, opioid, stimulant and tobacco use disorders, during the two years prior to pregnancy up to the day before the prenatal substance use screening date. Outcome variables were any cannabis use, frequency of self-reported cannabis use and CUD during early pregnancy. Psychiatric disorder prevalence ranged from 0.2% (psychotic) to 14.3% (anxiety), and SUD ranged from 0.3% (stimulant/opioid) to 3.8% (tobacco). Psychiatric disorders were associated with cannabis use and CUD, with the strongest association for any use found for bipolar disorder (adjusted odds ratio [aOR] = 2.83; 95% confidence interval [CI] = 2.53-3.17) and the strongest association for CUD found for psychotic disorders (aOR = 10.01, 95% CI = 6.52-15.37). SUDs were associated with cannabis use and CUD, with the strongest association for any use found for tobacco use disorder (aOR = 4.03, 95% CI = 3.82-4.24) and the strongest association for CUD found for stimulant use disorder (aOR = 21.99, 95% CI = 16.53-29.26). Anxiety, bipolar, depressive disorders and tobacco use disorder were associated with greater odds of daily than monthly or less cannabis use. Psychiatric disorders and substance use disorders appear to be associated with elevated odds of any and frequent cannabis use as well as cannabis use disorder during early pregnancy. In most cases, the associations with cannabis outcomes were stronger for substance use disorders than other psychiatric disorders.

Guidelines of the Brazilian Society of Surgical Oncology for anatomopathological, immunohistochemical, and molecular testing in female tumors.

Precision medicine has revolutionized oncology, providing more personalized diagnosis, treatment, and monitoring for patients with cancer. In the context of female-specific tumors, such as breast, ovarian, endometrial, and cervical cancer, proper tissue collection and handling are essential for obtaining tissue, immunohistochemical (IHC), and molecular data to guide therapeutic decisions. To establish guidelines for the collection and handling of tumor tissue, to enhance the quality of samples for histopathological, IHC, genomic, and molecular analyses. These guidelines are fundamental in informing therapeutic decisions in cancer treatment. The guidelines were developed by a multidisciplinary panel of renowned specialists between June 12, 2013 and February 12, 2024. Initially, the panel deliberated on critical and controversial topics related to conducting precision medicine studies focusing on female tumors. Subsequently, 22 pivotal topics were identified within the framework and assigned to groups. These groups reviewed relevant literature and drafted preliminary recommendations. Following this, the recommendations were reviewed by the coordinators and received unanimous approval. Finally, the groups made the final adjustments, classified the level of evidence, and ranked the recommendations. The collection of surgical samples requires minimum quality standards to enable histopathological, IHC, genomic, and molecular analyses. These analyses provide crucial data for informing therapeutic decisions, significantly impacting potential survival gains for patients with female tumors.

Parental illness work across the attention deficit hyperactivity disorder diagnostic journey.

The process of referral, assessment, and diagnosis of attention deficit hyperactivity disorder (ADHD) within the UK is often protracted. Given that parents are frequently the instigators of the diagnostic process, understanding the experience of parents is important. Drawing on findings from a longitudinal study, this article explores how the parental experience of the ADHD diagnostic journey includes three significant and distinct forms of 'illness work'. Twenty-one semi-structured serial interviews were conducted over a 2-year period with seven parents of children on the ADHD diagnostic journey in North East England. We present three significant forms of parental illness work: (1) The 'diagnostic quest', parental work recognising and fighting for their children's needs and selfhood, seeking diagnosis and engaging with systems, (2) 'self-biographical illness work', the personal parental biographical response to the diagnostic journey and (3) 'child biographical illness work and recontextualizing the child', parental biographical adjustment and recontextualisation of their children. We advance Rasmussen etal.'s (2021) model by demonstrating its usefulness in understanding how parents with a personal ADHD diagnosis experience biographical disruption or cohesion in response to their children's diagnosis. That a child's diagnosis leads parents with ADHD to experience a self-biographical cohesive or disruptive response is a unique and significant finding.

Next-generation sequencing technology in cancer

Next-generation sequencing (NGS) technology has revolutionized cancer research and treatment by enabling comprehensive analysis of genetic mutations, alterations, and expression profiles. It allows for the identification of cancer-driving mutations, helping in the development of targeted therapies. NGS provides detailed insights into tumor heterogeneity, resistance mechanisms, and clonal evolution. Its high-throughput capacity facilitates large-scale studies, improving our understanding of cancer genomics. By enabling personalized treatment plans based on individual genetic profiles, NGS holds promise for more effective and tailored cancer therapies. Early reviews on cancer genomics often lacked comprehensive coverage of emerging technologies. They missed in-depth analysis of NGS advancements, their impact on cancer research, and clinical applications. The review addresses this gap by reviving a thorough examination of NGS methods, their role in identifying genetic mutations, and their potential in personalized cancer treatment, thus providing essential insights into the evolving landscape of cancer genomics. The article covers the advancements in technology and bioinformatic approaches for NGS data analysis. It delves into NGS applications in research and diagnostics, particularly for solid cancer diagnosis. The review highlights specific cancer types, including hereditary breast cancer, melanoma, prostate cancer, thyroid cancer, lung cancer, and colorectal cancer. It explores NGS contribution in understanding the genetic basis of these cancers and its potential for enhancing personalized diagnosis and treatment strategies. This review rectifies early lacunas by providing a comprehensive and updated examination of NGS technology, addressing gaps in previous analyses and emphasizes bioinformatic approaches for NGS data analysis, crucial for interpreting vast genomic data accurately. The review meets the current need for a thorough understanding of NGS’s role in personalized cancer treatment and research.

The challenges in diagnosing hypertrophic cardiomyopathy in the presence of arterial hypertension: a clinical case

Aim. To describe a clinical case and analyze the challenges in diagnosing hypertrophic cardiomyopathy (HCM) in a patient with concomitant arterial hypertension (AH). Materials and methods. The article presents the clinical case of HCM in the patient with concomitant AH that was observed in an ambulatory setting in the Municipal Enterprise “Dnipropetrovsk Regional Clinical Center for Diagnostics and Treatment” of Dnipropetrovsk Regional Council. Results. A 66-year-old woman N., diagnosed by her family physician with coronary artery disease (CAD): stable angina FC II, grade 2 AH, chronic heart failure FC II NYHA, was referred to a cardiologist because of experiencing exertional dyspnea, chest pain, and uncontrolled blood pressure despite treatment compliance. Echocardiography identified concentric left ventricular hypertrophy with the left ventricular outflow tract (LVOT) obstruction (a mean gradient of 35 mmHg as per catheterization). Cardiac MRI confirmed the diagnosis and coronary angiography ruled out CAD. Adjustments to the treatment regimen, taking into account HCM with LVOT obstruction, effectively alleviated the patient’s symptoms and stabilized her blood pressure. Conclusions. It is especially important to follow the guidelines of AH management and perform echocardiography in all patients with high blood pressure, so as not to miss the signs of concomitant HCM, particularly with LVOT obstruction. In addition, in the case of HCM, it is necessary to timely detect, provide prevention and manage patients at risk for sudden cardiac death. Since HCM encompasses various diagnoses with different pathogenesis and distinct management, cardiac MRI, enzymatic or genetic testing may be needed according to guidelines. As HCM is a relatively common inherited cardiac disease, general practitioners could often encounter such patients in everyday clinical practice. Hence, they should have a certain suspicion of this diagnosis in persons with AH whose left ventricular mass meets the criteria for HCM.

Applicability of combined high-frequency and contrast-enhanced ultrasound in finger extensor tendon injuries: three case reports

BackgroundBy combining high-frequency and contrast-enhanced ultrasound (CEUS), the position of the severed end of a finger extensor tendon injury and the injury classification can be determined as part of a comprehensive preoperative evaluation in clinical practice. However, there have been no reports of high-frequency ultrasound combined with CEUS for the preoperative diagnosis of human finger extensor tendon injury.Cases presentationOne case of complete rupture of the extensor tendon was diagnosed by ultrasound, which was completely consistent with the surgery; one case of incomplete rupture was ultimately confirmed clinically; and one case of distal phalangeal bone base avulsion fracture with tendon contusion and missed diagnosis on the first radiographic examination was confirmed by follow-up radiographic examination.ConclusionsDifferent types of finger extensor tendon injuries exhibit distinctive contrast-enhanced ultrasonography findings. Combined high-frequency and contrast-enhanced ultrasound can accurately locate the position of the severed end of the finger extensor tendon injury before surgery while observing the contrast agent filling area to clarify injury classification, providing a reliable imaging basis for clinical practice and ultimately developing personalized diagnosis and treatment plans for patients to ensure minimal trauma and pain, as well as optimal treatment effects.

Human Case of Leptospirosis During a Canine Disease Outbreak - Wyoming, 2023.

Leptospirosis is a zoonotic bacterial disease spread through the urine of infected animals; the typical incubation period is 5-14 days. In approximately 90% of human cases, illness is asymptomatic or mild, characterized by fever, chills, myalgia, nausea, vomiting, diarrhea, headache, calf pain, and conjunctival suffusion, but severe illness can progress to multiorgan dysfunction and death. Although Wyoming is considered a low-risk area for leptospirosis because of its cold and semiarid climate, the Wyoming Department of Health was notified of a probable human case in August 2023, the first reported in the state since 1983. The patient had occupational exposure to dogs but did not report other risk factors. The same week that the human patient's illness began, public health authorities received notification of an increase in canine leptospirosis cases. Public health authorities investigated to determine potential sources of infection, identify additional cases, and recommend control measures. After public health outreach activities were implemented, canine vaccination practices changed substantially in the affected city: a survey conducted after the outbreak revealed that all responding veterinary clinics in the affected city were recommending the vaccine more frequently to dog owners and reporting higher levels of owner compliance with vaccination recommendations. Increased vaccination coverage offers protection from leptospirosis for both dogs and persons exposed to them. Leptospirosis should be considered in the differential diagnosis of persons with occupational exposure to animals and clinically compatible signs and symptoms, including fever, chills, myalgia, nausea, vomiting, diarrhea, headache, calf pain, and conjunctival suffusion, irrespective of geographic location.

A novel copy number variant in the murine Cdh23 gene gives rise to profound deafness and vestibular dysfunction.

Hearing loss is the most common congenital sensory deficit worldwide and exhibits high genetic heterogeneity, making molecular diagnoses elusive for most individuals. Detecting novel mutations that contribute to hearing loss is crucial to providing accurate personalized diagnoses, tailored interventions, and improving prognosis. Copy number variants (CNVs) are structural mutations that are understudied, potential contributors to hearing loss. Here, we present the Abnormal Wobbly Gait (AWG) mouse, the first documented mutant exhibiting waltzer-like locomotor dysfunction, hyperactivity, circling behaviour, and profound deafness caused by a spontaneous CNV deletion in cadherin 23 (Cdh23). We were unable to identify the causative mutation through a conventional whole-genome sequencing (WGS) and variant detection pipeline, but instead found a linked variant in hexokinase 1 (Hk1) that was insufficient to recapitulate the AWG phenotype when introduced into C57BL/6J mice using CRISPR-Cas9. Investigating nearby deafness-associated genes revealed a pronounced downregulation of Cdh23 mRNA and a complete absence of full-length CDH23 protein, which is critical for the development and maintenance of inner ear hair cells, in whole head extracts from AWG neonates. Manual inspection of WGS read depth plots of the Cdh23 locus revealed a putative 10.4kb genomic deletion of exons 11 and 12 that was validated by PCR and Sanger sequencing. This study underscores the imperative to refine variant detection strategies to permit identification of pathogenic CNVs easily missed by conventional variant calling to enhance diagnostic precision and ultimately improve clinical outcomes for individuals with genetically heterogenous disorders such as hearing loss.

Prevalence of Self-Reported COVID-19 Infection in Persons with Complicated Mild to Severe Traumatic Brain Injury (TBI): A TBI Model Systems Study.

To describe the incidence of self-reported COVID-19 history in a longitudinal cohort of individuals with complicated mild to severe traumatic brain injury (TBI) and describe demographic, injury and functional differences based on history of COVID-19 infection. Individuals with complicated mild to severe TBI aged 16 or older at time of injury who were enrolled in the TBI Model Systems longitudinal cohort study, completed a baseline or follow-up interview between October 1, 2021-March 31, 2023, and provided information about COVID-19 history and timing of COVID-19 infection was collected. Of the 3,627 individuals included in the analysis, 29.5% reported a history of COVID-19 infection. Those with reported COVID-19 history tended to be younger, not of a racial/ethnic minority background, and greater functional status at follow up based on the Glasgow Outcome Scale-Extended scale compared to those with no reported COVID-19 history (p < 0.05). Among those with COVID-19 history, 61.8% did not receive medical care, 27.6% received medical care but no hospitalization, and 10.5% were hospitalized. Of those hospitalized, 21.4% required ventilator use. Incidence of COVID-19 diagnosis and related hospitalization characteristics in persons with complicated mild to severe TBI was similar to national incidence between March 2020-2023. Secondary effects of the COVID-19 pandemic on persons with TBI require investigation.

Can life events predict first-time suicide attempts? A nationwide longitudinal study.

The prevention paradox describes circumstances in which the majority of cases with a suicide attempt come from a population of low or moderate risk, and only a few from a 'high-risk' group. The assumption is that a low base rate in combination with multiple causes makes it impossible to identify a high-risk group with all suicide attempts. The best way to study events such as first-time suicide attempts and their causes is to collect event history data. Administrative registers were used to identify a group at higher risk of suicidal behaviour within a population of six national birth cohorts (N = 300,000) born between 1980 and 1985 and followed from age 15 to 29 years. Estimation of risk parameters is based on the discrete-time logistic odds-ratio model. Lifetime prevalence was 4.5% for first-time suicide attempts. Family background and family child-rearing factors were predicative of later first-time suicide attempts. A young person's diagnosis with psychiatric or neurodevelopmental disorders (ADHD, anxiety, depression, PTSD), and being a victim of violence or sex offences contributed to the explanatory model. Contrary to the prevention paradox, results suggest that it is possible to identify a discrete high-risk group (<12%) among the population from whom two thirds of all first-time suicide attempts occur, but one third of observed suicide attempts derived from low- to moderate-risk groups. Findings confirm the need for a combined strategy of universal, targeted and indicated prevention approaches in policy development and in strategic and practice responses, and some promising prevention strategies are presented.

Artificial Intelligence for the Management of Breast Cancer: An Overview.

Breast cancer is a severe global health problem, and early detection, accurate diagnosis, and personalized treatment is the key to improving patient outcomes. Artificial intelligence (AI) and machine learning (ML) have emerged as promising breast cancer research and clinical practice tools in recent years. Various projects are underway in early detection, diagnosis, prognosis, drug discovery, advanced image analysis, precision medicine, predictive modeling, and personalized treatment planning using artificial intelligence and machine learning. These projects use different algorithms, including convolutional neural networks (CNNs), support vector machines (SVMs), decision trees, and deep learning methods, to analyze and improve different types of data, such as clinical, genomic, and imaging data for breast cancer management. The success of these projects has the potential to transform breast cancer care, and continued research and development in this area is likely to lead to more accurate and personalized breast cancer diagnosis, treatment, and outcomes.

Toward Enabling Cardiac Digital Twins of Myocardial Infarction Using Deep Computational Models for Inverse Inference.

Cardiac digital twins (CDTs) have the potential to offer individualized evaluation of cardiac function in a non-invasive manner, making them a promising approach for personalized diagnosis and treatment planning of myocardial infarction (MI). The inference of accurate myocardial tissue properties is crucial in creating a reliable CDT of MI. In this work, we investigate the feasibility of inferring myocardial tissue properties from the electrocardiogram (ECG) within a CDT platform. The platform integrates multi-modal data, such as cardiac MRI and ECG, to enhance the accuracy and reliability of the inferred tissue properties. We perform a sensitivity analysis based on computer simulations, systematically exploring the effects of infarct location, size, degree of transmurality, and electrical activity alteration on the simulated QRS complex of ECG, to establish the limits of the approach. We subsequently present a novel deep computational model, comprising a dual-branch variational autoencoder and an inference model, to infer infarct location and distribution from the simulated QRS. The proposed model achieves mean Dice scores of 0.457 ±0.317 and 0.302 ±0.273 for the inference of left ventricle scars and border zone, respectively. The sensitivity analysis enhances our understanding of the complex relationship between infarct characteristics and electrophysiological features. The in silico experimental results show that the model can effectively capture the relationship for the inverse inference, with promising potential for clinical application in the future. The code is available at https://github.com/lileitech/MI_inverse_inference.