Accuracy of an Artificial Intelligence-Based Model for Estimating Leftover Liquid Food in Hospitals: Validation Study.

BACKGROUND: An accurate evaluation of the nutritional status of malnourished hospitalized patients at a higher risk of complications, such as frailty or disability, is crucial. Visual methods of estimating food intake are popular for evaluating the nutritional status in clinical environments. However, from the perspective of accurate measurement, such methods are unreliable. OBJECTIVE: The accuracy of estimating leftover liquid food in hospitals using an artificial intelligence (AI)-based model was compared to that of visual estimation. METHODS: The accuracy of the AI-based model (AI estimation) was compared to that of the visual estimation method for thin rice gruel as staple food and fermented milk and peach juice as side dishes. A total of 576 images of liquid food (432 images of thin rice gruel, 72 of fermented milk, and 72 of peach juice) were used. The mean absolute error, root mean squared error, and coefficient of determination (R2) were used as metrics for determining the accuracy of the evaluation process. Welch t test and the confusion matrix were used to examine the difference of mean absolute error between AI and visual estimation. RESULTS: The mean absolute errors obtained through the AI estimation approach were 0.63 for fermented milk, 0.25 for peach juice, and 0.85 for the total. These were significantly smaller than those obtained using the visual estimation approach, which were 1.40 (P<.001) for fermented milk, 0.90 (P<.001) for peach juice, and 1.03 (P=.009) for the total. By contrast, the mean absolute error for thin rice gruel obtained using the AI estimation method (0.99) did not differ significantly from that obtained using visual estimation (0.99). The confusion matrix for thin rice gruel showed variation in the distribution of errors, indicating that the errors in the AI estimation were biased toward the case of many leftovers. The mean squared error for all liquid foods tended to be smaller for the AI estimation than for the visual estimation. Additionally, the coefficient of determination (R2) for fermented milk and peach juice tended to be larger for the AI estimation than for the visual estimation, and the R2 value for the total was equal in terms of accuracy between the AI and visual estimations. CONCLUSIONS: The AI estimation approach achieved a smaller mean absolute error and root mean squared error and a larger coefficient of determination (R2) than the visual estimation approach for the side dishes. Additionally, the AI estimation approach achieved a smaller mean absolute error and root mean squared error compared to the visual estimation method, and the coefficient of determination (R2) was similar to that of the visual estimation method for the total. AI estimation measures liquid food intake in hospitals more precisely than visual estimation, but its accuracy in estimating staple food leftovers requires improvement.

Hyponatremia is a Prognostic Factor in Patients Receiving Nutrition Support.

BACKGROUND: Hyponatremia, the most common electrolyte disorder, has been reported to be related to increased mortality. However, the association between hyponatremia and prognoses remains unclear in patients with nutrition support team (NST) intervention. This study aimed to determine the prevalence of abnormal serum sodium levels, its relation to patient data, and the impact of hyponatremia on prognosis. METHODS: Patients who received nutrition support at Tokushima University Hospital for the first time and whose serum sodium levels were measured at the start of NST intervention were enrolled. Patients were classified into three groups according to their serum Na levels at the start of NST intervention: hyponatremia group, normonatremia group, and hypernatremia group. RESULTS: In the hyponatremia group compared to the normonatremia group, body weight and body mass index were significantly lower. C-reactive protein levels and urea nitrogen/creatinine ratios were significantly higher. Meanwhile, there was no significant difference in the estimated glomerular filtration rate among the groups. The prevalence of malnutrition and anemia were the highest in the hyponatremia group. The 3-year survival rate was approximately 45% in the hyponatremia group, which was the lowest of all three groups. The mortality risk ratio of the hyponatremia group to the normonatremia group was 2.29. CONCLUSIONS: Hyponatremia in NST intervention patients is an independent prognostic predictor. Therefore, adding an assessment of serum sodium at the beginning of NST intervention can identify patients at high risk at an early stage and may improve the quality of NST activity.

Association of increased renal Cyp24a1 gene expression with low plasma 1,25-dihydroxyvitamin D levels in rats with streptozotocin-induced diabetes.

Decreases in plasma vitamin D concentrations have been reported in diabetes, although the mechanism involved in this decrease is unclear. Here, we investigated the association between Cyp24a1, a vitamin D catabolic enzyme, and abnormalities in vitamin D metabolism in streptozotocin-induced diabetes rats, an animal model of type 1 diabetes. Plasma 1,25-dihydroxyvitamin D [1,25(OH)2D] levels were significantly lower in streptozotocin-induced diabetes rats and renal Cyp24a1 mRNA expression levels were increased. Western blotting analysis of streptozotocin-induced diabetes rats kidney tissues with anti-CYP24A1 antibody showed a strong signal around 40 kDa, which differs from the predicted 50-55 kDa molecular weight for full-length Cyp24a1 and could represent the Cyp24a1-splicing variant that lacks exons 1 and 2. We observed high levels of renal Cyp24a1-splicing variant mRNA expression in streptozotocin-induced diabetes rats. We also confirmed transcriptional up-regulation of endogenous Cyp24a1 mRNA expression through glucocorticoid receptors by glucocorticoid in opossum kidney proximal cells. Taken together, our results indicated that high Cyp24a1 expression levels may play a role in the decrease of plasma 1,25(OH)2D levels in streptozotocin-induced diabetes rats. High plasma corticosterone levels in diabetes may affect transcriptional regulation to promote increases in Cyp24a1 expression.

Sterol regulatory element binding protein 1 trans-activates 25-hydroxy vitamin D3 24-hydroxylase gene expression in renal proximal tubular cells.

The physiological activity of the steroid derived hormone vitamin D is regulated by several enzymatic steps. Both 25-hydroxy vitamin D3 1α-hydroxylase (CYP27B1) and 25-hydroxyvitamin D3 24-hydroxylase (CYP24A1) modulate blood levels of 1,25-dihydroxyvitamin D3, an activated form of vitamin D. We previously demonstrated that CYP27B1 expression was trans-activated by sterol regulatory element binding protein 1 (SREBP1), although whether SREBP1 also regulates CYP24A1 transcription was unclear. Here we investigated the ability of SREBP1 to affect CYP24A1 transcription. In a luciferase reporter assay, mouse and human CYP24A1 promoter activity was strongly activated by SREBP1 in opossum kidney proximal tubular cells (OK-P). Three putative SREs (pSREs) were found in the mouse Cyp24a1 gene promoter and the SREBP1 protein showed specific binding to the pSRE1 element in EMSAs. Site-directed mutagenesis of the pSRE1 element strongly decreased SREBP1-mediated Cyp24a1 gene transcription. Moreover, siRNA-mediated SREBP1 knock-down repressed CYP24A1 expression in human renal proximal tubular epithelial cells (HKC-8). In animal studies, mice given various doses of thyroid hormone (T3) showed dose-dependent reductions in renal Srebp1c and Cyp24a1 mRNA levels. Taken together, our results suggest that SREBP1 trans-activates CYP24A1 expression through SREBP binding elements present in the promoter.

The age-related changes of dietary phosphate responsiveness in plasma 1,25-dihydroxyvitamin D levels and renal Cyp27b1 and Cyp24a1 gene expression is associated with renal α-Klotho gene expression in mice.

In this study, we investigated the relationship between age-related changes in renal α-Klotho gene expression, vitamin D metabolism and the responsiveness of dietary phosphate in 1, 2 and 13 month-old mice fed a high phosphate (phosphate 1.2%) diet or low phosphate (phosphate 0.02%) diet for 5 days. We found that 1,25-dihydroxyvitamin D levels in plasma were significantly lower in the high phosphate group than the low phosphate group for 1 and 2 month-old mice, but not 13 month-old mice. In addition, in the high phosphate group plasma 1,25-dihydroxyvitamin D levels were decreased in 2 month-old mice relative to 1 month-old mice, but 13 month-old mice had higher levels than 2 month-old mice. In fact, plasma 1,25-dihydroxyvitamin D levels showed a significant correlation with vitamin D metabolism gene Cyp27b1 and Cyp24a1 mRNA expression in the high phosphate group. Interestingly, renal α-Klotho mRNA and protein levels were significant change with age. Furthermore, α-Klotho mRNA expression showed a significant negative correlation with plasma 1,25-dihydroxyvitamin D levels in the high phosphate group. Our results suggest that age-related alterations in renal α-Klotho expression could affect the responsiveness of dietary phosphate to vitamin D metabolism.