Machine learning to predict the occurrence of thyroid nodules: towards a quantitative approach for judicious utilization of thyroid ultrasonography.

Introduction: Ultrasound is instrumental in the early detection of thyroid nodules, which is crucial for appropriate management and favorable outcomes. However, there is a lack of clinical guidelines for the judicious use of thyroid ultrasonography in routine screening. Machine learning (ML) has been increasingly used on big data to predict clinical outcomes. This study aims to leverage the ML approach in assessing the risk of thyroid nodules based on common clinical features. Methods: Data were sourced from a Chinese cohort undergoing routine physical examinations including thyroid ultrasonography between 2013 and 2023. Models were established to predict the 3-year risk of thyroid nodules based on patients' baseline characteristics and laboratory tests. Four ML algorithms, including logistic regression, random forest, extreme gradient boosting, and light gradient boosting machine, were trained and tested using fivefold cross-validation. The importance of each feature was measured by the permutation score. A nomogram was established to facilitate risk assessment in the clinical settings. Results: The final dataset comprised 4,386 eligible subjects. Thyroid nodules were detected in 54.8% (n=2,404) individuals within the 3-year observation period. All ML models significantly outperformed the baseline regression model, successfully predicting the occurrence of thyroid nodules in approximately two-thirds of individuals. Age, high-density lipoprotein, fasting blood glucose and creatinine levels exhibited the highest impact on the outcome in these models. The nomogram showed consistency and validity, providing greater net benefits for clinical decision-making than other strategies. Conclusion: This study demonstrates the viability of an ML-based approach in predicting the occurrence of thyroid nodules. The findings highlight the potential of ML models in identifying high-risk individuals for personalized screening, thereby guiding the judicious use of ultrasound in this context.

Clinical observation of Yiqi Qingdu prescription on the treatment of intermediate-stage and advanced non-small-cell lung cancer.

OBJECTIVE: To observe the effects of the Yiqi Qingdu prescription () on intermediate-stage and advanced non-small-cell lung cancer (NSCLC). METHODS: In total, 300 patients with intermediate-stage or advanced NSCLC were randomly and equally divided into three groups using computer-generated random numbers as follows: Western medicine (WM), Chinese medicine (CM), and integrated Traditional Chinese and Western Medicine (IM). After 3 months of treatment, the overall response rate (ORR); disease control rate (DCR); symptom score (SS); Karnofsky performance status (KPS); adverse event score; counts of CD3 + , CD4 + , and CD8 + cells; CD4 + /CD8 + ratio; and carcinoembryonic antigen (CEA) level were compared among the groups. RESULTS: The ORRs were 30.36% , 20.24% , and 7.87% in the IM, CM, and WM groups, respectively, whereas the DCRs were 85%, 75%, and 73%, respectively. Compared to the CM group, the ORR was significantly higher in the WM and IM groups, whereas the DCR was significantly higher in the IM group (all P < 0.05). SS was obviously higher in the WM group than in the other two groups (both P < 0.01). KPS was significantly lower in the WM group after treatment (P = 0.005). The mean number of adverse events was significantly lower in the CM (2.2 ± 1.3) and IM (2.4 ± 1.3) groups than in the WM group (4.6 ± 1.7, both P < 0.05). CD3 + cell counts were significantly decreased in the WM group (P = 0.031). In the IM group, CD8+ cell counts were increased after treatment, whereas the CD4 + /CD8 + ratio was decreased (both P < 0.01). Compared with the WM group, CD3 + (P = 0.01), CD4 + (P = 0.044), and CD8 + (P = 0.009) cell counts were significantly higher in the IM group, whereas the CD4+ /CD8+ ratio was significantly lower (P = 0.011). Relative to the CM group, CD8 + cell counts were significantly higher (P = 0.001) and the CD4+ /CD8+ ratio was significantly lower in the IM group (P = 0.001). CEA levels were significantly increased in the CM group (P = 0.023). CONCLUSION: The Yiqi Qingdu prescription can improve the outcomes of WM in patients with NSCLC.

Association of Changes in Metabolic Syndrome Status With the Incidence of Thyroid Nodules: A Prospective Study in Chinese Adults.

Although several cross-sectional studies have shown an association of metabolic syndrome (MetS) with nodular thyroid disease, related prospective studies are scarce. This study investigated the association of MetS with thyroid nodule (TN) incidence in Chinese adults, and explored whether the development of or recovery from MetS is associated with changes in the risk of developing TNs. A total of 4,749 Chinese aged 18-65 years were involved in this 6-year prospective study. The association of MetS with TN prevalence was examined. TN-free individuals at baseline (n = 3,133) were further examined. TN incidence rates in groups with different MetS statuses (MetS-free, MetS-developed, MetS-recovery and MetS-chronic) were analyzed. Of all participants, 18.21 and 31.65% had MetS and TNs, respectively. MetS patients had a higher TN prevalence than the non-MetS group (31.08 vs. 19.81% in males, p < 0.01; 59.52 vs. 39.59% in females, p < 0.01). Sex, age and MetS were independent risk factors for TNs. At a median follow up of 5.94 years, the MetS-chronic group (4.37/100 person-years) had a higher risk of TNs (adjusted incidence rate ratio [IRR] = 1.288 [95% CI 1.014-1.636]) compared with the MetS-free group (2.72/100 person-years) in the whole cohort. In males, the MetS-chronic group (3.76/100 person-years) had a higher risk of TNs (adjusted IRR = 1.367 [95% CI 1.017-1.835]) compared with the MetS-free group (2.31/100 person-years). In females, the risk of TNs was significantly higher in the MetS-chronic (6.44/100 person-years) and MetS-developed (6.31/100 person-years) groups compared with the MetS-free group (3.23/100 person-years).

Lipopolysaccharide-enhanced early proliferation of insulin secreting NIT-1 cell is associated with nuclear factor-kappaB- mediated inhibition of caspase 3 cleavage.

BACKGROUND: Increased levels of plasma lipopolysaccharide (LPS) have been found in obesity and diabetes patients. This study was to investigate the effect of LPS on pancreatic beta-cell viability and the involvement of caspase 3 in NIT-1 cell line. METHODS: Mouse insulinoma NIT-1 cells were treated with LPS for the indicated time and dose. Cell viability was measured by cell counting kit-8 reagent. Toll-like receptor 4 (TLR4), caspase 3 and cleaved caspase 3 were detected by Western blotting. Insulin was determined by radioimmunoassay (RIA). RESULTS: LPS promoted NIT-1 cell proliferation at 1 µg/ml, peaked at 72 hours of incubation. A reduction in cleavage of caspase 3 was observed upon LPS treatment. Bay11-7082, a specific inhibitor of nuclear factor (NF)-κB, blunted LPS-induced inhibition of caspase 3 cleavage. Reduction in chronic insulin secretion was observed after treatment with LPS at 1 µg/ml for 48 and 72 hours, not for 24 hours. TLR4 protein was upregulated when NIT-1 cells were treated with LPS at 1 µg/ml for 24 hours. CONCLUSIONS: LPS promotes early NIT-1 cell proliferation in association with NF-κB-mediated inhibition of caspase 3 cleavage. LPS exerts a time-dependent inhibitory effect on chronic insulin secretion from NIT-1 cells.