Neural networks in the predictive diagnosis of cognitive impairment in type 1 and type 2 diabetes mellitus
Notice bibliographique
Résumé
BACKGROUND: Cognitive dysfunction, including mild cognitive impairment and dementia, is increasingly recognized as a serious complication of diabetes mellitus (DM) that affects patient well-being and disease management. Magnetic resonance imaging (MRI)-studies have shown varying degrees of cortical atrophy, cerebral infarcts, and deep white matter lesions. To explain the relationship between DM and cognitive decline, several hypotheses have been proposed, based on the variability of glycemia leading to morphometric changes in the brain. The ability to predict cognitive decline even before its clinical development will allow the early prevention of this pathology, as well as to predict the course of the existing pathology and to adjust medication regimens. AIM: To create a computer neural network model for predicting the development of cognitive impairment in DM on the basis of brain neuroimaging techniques. MATERIALS AND METHODS: The study was performed in accordance with the standards of good clinical practice; the protocol was approved by the Ethics Committee. The study included 85 patients with type 1 diabetes and 95 patients with type 2 diabetes, who were divided into a group of patients with normal cognitive function and a group with cognitive impairment. The patient groups were comparable in age and duration of disease. Cognitive impairment was screened using the Montreal Cognitive Assessment Scale. Data for glycemic variability were obtained using continuous glucose monitoring (iPro2, Libre). A standard MRI scan of the brain was performed axially, sagittally, and coronally on a Signa Creator E, GE Healthcare, 1.5 Tesla, China. For MRI data processing we used Free Surfer program (USA) for analysis and visualization of structural and functional neuroimaging data from cross-sectional or longitudinal studies, and for segmentation we used Recon-all batch program directly. All statistical analyses and data processing were performed using Statistica Statsofi software (version 10) on Windows 7/XP Pro operating systems. The IBM WATSON cognitive system was used to build a neural network model. RESULTS: As a result of the study, cognitive impairment in DM type 1was predominantly of mild degree 36.9% (n=24) and moderate degree 30.76% (n=20), and in DM type 2 mild degree 37% (n=30), moderate degree 49.4% (n=40) and severe degree 13.6% (n=11). Cognitive functions in DM type 1 were impaired in memory and attention, whereas in DM type 2 they were also impaired in tasks of visual-constructive skills, fluency, and abstraction (p0.001). The analysis revealed differences in glycemic variability indices in patients with type 1 and type 2 DM and cognitive impairment. Standard MRI of the brain recorded the presence of white and gray matter changes (gliosis and leukoareosis). General and regional cerebral atrophy is characteristic of type 1 and type 2 DM, which is associated with dysglycemia. When building neural network models for type 1 diabetes, the parameters of decreased volumes of the brain regions determine the development of cognitive impairment by 93.5%, whereas additionally, the coefficients of glycemic variability by 98.5%. The same peculiarity was revealed in type 2 DM 95.3% and 97.9%, respectively. CONCLUSION: In DM type 1 and type 2 with cognitive impairment, elevated coefficients of glycemic variability are more frequently recorded. This publication describes laboratory and instrumental parameters as potential diagnostic options for effective management of DM and prevention of cognitive impairment. Neural network models using glycemic variability coefficients and MR morphometry allow for predictive diagnosis of cognitive disorders in both types of diabetes.
Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.
Comment cette classification a été obtenuedéplier
Prédiction distillée sur la base complète
Imitation des enseignantsNi prévalence calibrée, ni vérité terrain. Validation humaine à venir. Apprise à partir de 10 348 étiquettes directes de Codex et de 10 348 étiquettes directes de Gemma. Le mode candidate est l'union des têtes enseignantes seuillées; le consensus est leur intersection. Ces sorties portent le statut machine_predicted_unvalidated et ne sont ni des étiquettes humaines ni des étiquettes directes de modèles de pointe.
Scores Codex et Gemma par catégorie
| Catégorie | Codex | Gemma |
|---|---|---|
| Métarecherche | 0,000 | 0,000 |
| Méta-épidémiologie (sens strict) | 0,000 | 0,000 |
| Méta-épidémiologie (sens large) | 0,000 | 0,000 |
| Bibliométrie | 0,000 | 0,001 |
| Études des sciences et des technologies | 0,000 | 0,000 |
| Communication savante | 0,000 | 0,000 |
| Science ouverte | 0,000 | 0,000 |
| Intégrité de la recherche | 0,000 | 0,000 |
| Charge utile insuffisante (le modèle a refusé de juger) | 0,000 | 0,000 |
Scores machine (provisoires)
Les deux têtes enseignantes du modèle étudiant, lues sur ce travail. Un score ordonne la base pour la relecture; il n'affirme jamais une catégorie, et le statut de validation accompagne chaque rangée tel quel.
Scores de référence d'un modèle non mature (critères de maturité non atteints, 7 itérations). Un score ordonne; il n'affirme jamais une catégorie.
score_only:v0-immature-baseline · tel quel depuis la passe de notation : score_only signifie que le nombre peut ordonner les travaux, et qu'aucune étiquette de catégorie n'en découleClassification
machine, non validéePrédiction automatique; un appel candidat d’une seule tête enseignante, pas un consensus.
Le détail, modèle par modèle et score par score, se trouve en fin de page sous « Comment cette classification a été obtenue ».