Prognosticating the outcome of intensive care in older patients—a narrative review

Annals of Intensive Care, Journal Year: 2024, Volume and Issue: 14(1)

Published: June 22, 2024

Abstract Prognosis determines major decisions regarding treatment for critically ill patients. Statistical models have been developed to predict the probability of survival and other outcomes intensive care. Although they were trained on characteristics large patient cohorts, often do not represent very old patients (age ≥ 80 years) appropriately. Moreover, heterogeneity within this particular group impairs utility statistical predictions informing decision-making in individuals. In addition these methodological problems, diversity cultural attitudes, available resources as well variations legal professional norms limit generalisability prediction models, especially with complex multi-morbidity pre-existing functional impairments. Thus, current approaches prognosticating are imperfect can generate substantial uncertainty about optimal trajectories critical care individual. This article presents state art new predicting Special emphasis has given integration into individual requires quantification prognostic a careful alignment preferences patients, who might prioritise over survival. Since performance outcome may improve time, time-limited trials be an appropriate way increase confidence life-sustaining treatment.

Language: Английский

---

A Deep Learning-Based Framework Oriented to Pathological Gait Recognition with Inertial Sensors

Sensors, Journal Year: 2025, Volume and Issue: 25(1), P. 260 - 260

Published: Jan. 5, 2025

Abnormal locomotor patterns may occur in case of either motor damages or neurological conditions, thus potentially jeopardizing an individual’s safety. Pathological gait recognition (PGR) is a research field that aims to discriminate among different walking patterns. A PGR-oriented system benefit from the simulation disorders by healthy subjects, since acquisition actual pathological gaits would require higher experimental time larger sample size. Only few works have exploited abnormal patterns, emulated unimpaired individuals, perform PGR with Deep Learning-based models. In this article, authors present workflow based on convolutional neural networks recognize normal and behaviors means inertial data related nineteen subjects. Although preliminary feasibility study, its promising performance terms accuracy computational pave way for more realistic validation data. light this, classification outcomes could support clinicians early detection tracking rehabilitation advances real time.

Language: Английский

---

Event-related potential markers of subjective cognitive decline and mild cognitive impairment during a sustained visuo-attentive task

NeuroImage Clinical, Journal Year: 2025, Volume and Issue: 45, P. 103760 - 103760

Published: Jan. 1, 2025

Language: Английский

---

Advancements in deep learning for early diagnosis of Alzheimer’s disease using multimodal neuroimaging: challenges and future directions

Frontiers in Neuroinformatics, Journal Year: 2025, Volume and Issue: 19

Published: May 2, 2025

Introduction Alzheimer’s disease is a progressive neurodegenerative disorder challenging early diagnosis and treatment. Recent advancements in deep learning algorithms applied to multimodal brain imaging offer promising solutions for improving diagnostic accuracy predicting progression. Method This narrative review synthesizes current literature on applications using neuroimaging. The process involved comprehensive search of relevant databases (PubMed, Embase, Google Scholar ClinicalTrials.gov ), selection pertinent studies, critical analysis findings. We employed best-evidence approach, prioritizing high-quality studies identifying consistent patterns across the literature. Results Deep architectures, including convolutional neural networks, recurrent transformer-based models, have shown remarkable potential analyzing neuroimaging data. These models can effectively structural functional modalities, extracting features associated with pathology. Integration multiple modalities has demonstrated improved compared single-modality approaches. also promise predictive modeling, biomarkers forecasting Discussion While approaches show great potential, several challenges remain. Data heterogeneity, small sample sizes, limited generalizability diverse populations are significant hurdles. clinical translation these requires careful consideration interpretability, transparency, ethical implications. future AI neurodiagnostics looks promising, personalized treatment strategies.

Language: Английский

---

Event-Related Potential Markers of Subject Cognitive Decline and Mild Cognitive Impairment during a sustained visuo-attentive task

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: Jan. 30, 2024

Abstract Subjective cognitive decline (SCD), mild impairment (MCI), or severe Alzheimer’s disease stages are still lacking clear electrophysiological correlates. In 178 individuals (119 SCD, 40 MCI, and 19 healthy subjects (HS)), we analysed event-related potentials recorded during a sustained visual attention task, aiming to distinguish biomarkers associated with clinical conditions task performance. We observed condition-specific anomalies in (ERPs) encoding (P1/N1/P2) decision-making (P300/P600/P900): SCD showed attenuated dynamics compared HS, while MCI amplified dynamics, except for P300, which matched severity. ERP features confirmed non-monotonic trend, showing higher neural resource recruitment. Moreover, performance correlated gain latencies across early late components. These findings enhanced the understanding of mechanisms underlying suggested potential diagnosis intervention. Highlights decision (P600/P900) ERPs, exhibited SCD. P300 demonstrated recruitment resources, indicating trend between conditions. Task multiple

Language: Английский

---

Artificial Intelligence in Newborn Medicine

Newborn, Journal Year: 2024, Volume and Issue: 3(2), P. 96 - 110

Published: June 21, 2024

The development of artificial intelligence (AI) algorithms has evoked a mixed-feeling reaction, combination excitement but also some trepidation, with reminders caution coming up each time novel AI-related academic/medical software program is proposed.There awareness, hesitancy, that these could turn out to be continuous, transformational source clinical and educational information.Several AI varying strengths weaknesses are known, the deep-learning pathways known as Generative Pre-trained Transformers (GPT) have most interest decision-support systems.Again, tools still need validation all steps should undergo multiple checks cross-checks prior any implementation in human medicine.If, however, testing eventually confirms utility pathways, there possibility non-incremental advancement immense value.Artificial can helpful by facilitating appropriate analysis large bodies data available not being uniformly comprehensively analyzed at centers.It promote appropriate, timely diagnoses, for efficacy less bias, fewer diagnostic medication errors, good follow-up.Predictive modeling help allocation resources identifying at-risk newborns right outset.Artificial develop information packets engage educate families.In academics, it an unbiased, allinclusive medical literature on continuous basis education research.We know will challenges protection privacy handling data, bias algorithms, regulatory compliance.Continued efforts needed understand streamline AI.However, if community hesitates today overseeing this juggernaut, inclusion (or not) medicine might stop-it just gradually get extrapolated into patient care from other organizations/industry cost reasons, justification based actual data.If we do involved process oversee development/incorporation newborn medicine, questions making decisions change who, which, when, how.Maybe scenario.To conclude, definite benefits; embrace developments valuable assist physicians analyzing complex datasets, which facilitate identification key facts/findings missed studied humans.On hand, well-designed critical expert review board mandatory prevent AI-generated systematic errors.

Language: Английский

---