Machine learning prediction model of major adverse outcomes after pediatric congenital heart surgery: a retrospective cohort study.

BACKGROUND: Major adverse postoperative outcomes (APOs) can greatly affect mortality, hospital stay, care management and planning, and quality of life. This study aimed to evaluate the performance of five machine learning (ML) algorithms for predicting four major APOs after pediatric congenital heart surgery and their clinically meaningful model interpretations. METHODS: Between August 2014 and December 2021, 23 000 consecutive pediatric patients receiving congenital heart surgery were enrolled. Based on the split date of 1 January 2019, the authors selected 13 927 participants for the training cohort, and 9073 participants for the testing cohort. Four predefined major APOs including low cardiac output syndrome (LCOS), pneumonia, renal failure, and deep venous thrombosis (DVT) were investigated. Thirty-nine clinical and laboratory features were inputted in five ML models: light gradient boosting machine (LightGBM), logistic regression (LR), support vector machine, random forest, and CatBoost. The performance and interpretations of ML models were evaluated using the area under the receiver operating characteristic curve (AUC) and Shapley Additive Explanations (SHAP). RESULTS: In the training cohort, CatBoost algorithms outperformed others with the mean AUCs of 0.908 for LCOS and 0.957 for renal failure, while LightGBM and LR achieved the best mean AUCs of 0.886 for pneumonia and 0.942 for DVT, respectively. In the testing cohort, the best-performing ML model for each major APOs with the following mean AUCs: LCOS (LightGBM), 0.893 (95% CI: 0.884-0.895); pneumonia (LR), 0.929 (95% CI: 0.926-0.931); renal failure (LightGBM), 0.963 (95% CI: 0.947-0.979), and DVT (LightGBM), 0.970 (95% CI: 0.953-0.982). The performance of ML models using only clinical variables was slightly lower than those using combined data, with the mean AUCs of 0.873 for LCOS, 0.894 for pneumonia, 0.953 for renal failure, and 0.933 for DVT. The SHAP showed that mechanical ventilation time was the most important contributor of four major APOs. CONCLUSIONS: In pediatric congenital heart surgery, the established ML model can accurately predict the risk of four major APOs, providing reliable interpretations for high-risk contributor identification and informed clinical decisions-making.

Comparison of Femoral Triangle Block in Combination with IPACK to Local Periarticular Injection in Total Knee Arthroplasty.

OBJECTIVE: This article studied the efficacy of two different analgesic methods after unilateral primary total knee arthroplasty (TKA) to find an effective analgesic method. METHODS: A randomized, double-blind, placebo, parallel, and controlled study was performed to evaluate the benefits of combining the femoral triangle block (FTB) and the interspace between the popliteal artery and the capsule of the posterior knee (IPACK). Forty patients diagnosed with knee osteoarthritis and underwent unilateral primary TKA with FTB and IPACK were divided grouped into the experimental group, and 40 patients undergoing TKA with intra-articular cocktail analgesic mixture local injection were grouped into the control group. All patients received the patient-controlled anesthesia pump for analgesia at postoperative 48 hours. The main indexes were postoperative knee joint rest and activity pain (visual analog scale) and muscle strength of the affected limb; secondary indexes were anesthetic consumption, total morphine consumption, range of motion, and complications (such as postoperative nausea and vomiting [PONV]). RESULTS: There was no significant difference in the general data of each treatment group. Compared with the conventional group, the quadriceps muscle strength of the combined FTB and IPACK group was higher with significant statistical differences after surgery (p < 0.05). At postoperative 2, 6, 12, 24, 48, and 72 hours, active pain was better than in the conventional group (p < 0.05). Resting pain was significantly smaller than the traditional group only at postoperative 2, 6, 12, and 48 hours (p < 0.05). Morphine consumption, anesthetics consumption, and hospitalization time were lower than the conventional group, the difference being statistically significant. There were no significant differences between the two groups in postoperative wound healing, infection incidence, blood pressure, heart rate, rash, respiratory depression, deep vein thrombosis, and urinary retention. There were also no significant differences in PONV (p > 0.05). CONCLUSION: Combining FTB and IPACK significantly increased the quadriceps muscle in patients, together with relieving early pain and reducing the amount of anesthetic consumption at different postoperative intervals.

Early Efficacy of Type I Collagen-Based Matrix-Assisted Autologous Chondrocyte Transplantation for the Treatment of Articular Cartilage Lesions.

Background: Articular cartilage is a complex structure that allows for low frictional gliding and effective shock absorption. Various sports injuries and inflammatory conditions can lead to lesions in the articular cartilage, which has limited regenerative potential. Type I collagen combined with autologous chondrocytes in a three-dimensional culture were used to induce the regeneration of single-layer autologous expanded chondrocytes without chondrogenic differentiation. Purpose: To assess the clinical, radiological, and histological changes following collagen-based autologous chondrocyte transplantation (MACT) for chondral knee lesions. Methods: The study prospectively enrolled 20 patients with symptomatic knee chondral lesions (mean size lesion was 2.41 ± 0.43 cm2, range: 2.0-3.4 cm2) in the lateral femoral condyle and femoral groove who underwent type I collagen-based MACT between July 2017 and July 2019. knee injury and osteoarthritis outcome score (KOOS) was assessed before the procedure, and periodic clinical follow-up was conducted every 3 months for a maximum of 12 months following the procedure and at 1-year intervals thereafter. Magnetic resonance imaging (MRI) T2 mapping of repaired cartilage was also used for the quantitative analysis of regeneration. In one patient, second-look arthroscopy was performed to assess cartilage regeneration characteristics, and a portion of regenerated cartilage was harvested for histological evaluation 12 months after implantation. Results: At pre-operation and at three, six, 12, and 24 months after the operation, KOOS pain, symptoms, daily life activities, sports and recreation, as well as the quality of life were significantly improved between every two time points. Hematoxylin and eosin (HE) staining indicated that the newly formed cartilage was comprised of naive chondrocytes. Safranin O-fast (S-O) green staining of the regenerated tissue revealed fibroblast-like cells surrounded by glycosaminoglycans. Immunohistochemistry (IHC) analysis indicated that collagen type II was uniformly distributed at the deep zone of articular cartilage and type I collagen mainly depositing in the superficial cartilage layer. The T2 values for repaired tissue gradually decreased, eventually approaching near-average values. Conclusion: The present study demonstrated that type I collagen-based MACT is a clinically effective treatment for improving functionality and pain levels. Histological evidence confirmed hyaline cartilage induction and showed that repaired cartilage tended to emerge from the deep to the superficial layer. The quantitative MRI T2 mapping test indicated that there still was a difference between the transplanted cartilage and the surrounding hyaline cartilage. Taken together, the current method represents an efficient approach for the restoration of knee cartilage lesions.

NIR spectroscopy coupled with chemometric algorithms for the prediction of cadmium content in rice samples.

Fast determination of heavy metals is necessary and important to ensure the safety of crops. The potential of near-infrared spectroscopy coupled with chemometric technology for quantitative analysis of cadmium in rice was investigated. A total of 825 rice samples were collected and scanned by NIRS. The Kennard-Stone method was applied to divide the samples into calibration and validation sets. Before modeling, the spectrum was preprocessed using first derivation to reduce the baseline shift. Different chemometric tools such as interval partial least squares, moving window partial least squares, synergy interval partial least squares, and backward interval partial least squares were proposed to extract and optimize spectral interval from full-spectrum data. The performance of the calibration models generated on the basis of different regression algorithms was compared and evaluated. Results showed that the PLS models based on four chemometric algorithms outperformed the full-spectrum PLS model. Among the tools, biPLS performed better with the optimal subinterval selection. The root-mean-square error of prediction and correlation coefficient (R) of the biPLS model were 0.2133 and 0.9020, respectively. In addition, the low root-mean-square error of cross-validation was obtained in biPLS, which was 0.1756. NIRS technology combined with biPLS could be considered as an effective and convenient tool for primary screening and measuring of cadmium content in rice. In comparison with classical methodologies, this new technology was beneficial because of its eco-friendliness, fast analysis, and virtually no sample preparation required.

miR-107 affects cartilage matrix degradation in the pathogenesis of knee osteoarthritis by regulating caspase-1.

BACKGROUND: Knee osteoarthritis (KOA) seriously affects the quality of life of KOA patients. This study aimed to investigate whether miR-107 could regulate KOA through pyroptosis to affect collagen protein secreted by chondrocytes through IL-1ß. METHODS: The proliferation of chondrocytes was detected by CCK-8 assay. RT-qPCR analysis was used to identify miR-107 expression and transfection effects. The expression of Col II, IL-1ß, IL-18, and MMP13 in supernatant of chondrocytes or chondrocytes was detected by ELISA assay and western blot analysis. The pyroptosis of chondrocytes was analyzed by TUNEL assay and the expression of pyroptosis-related proteins was analyzed by western blot. Luciferase reporter assay confirmed the relation of miR-107 to caspase-1. RESULTS: The proliferation of chondrocytes was decreased after LPS induction and further decreased by treatment of ATP. Single LPS treatment for chondrocytes downregulated the Col II expression while upregulated the expression of IL-1ß, IL-18, and MMP-13, which was further changed by ATP treatment. miR-107 expression was decreased in chondrocytes induced by LPS and further decreased in chondrocytes induced by LPS and ATP. In addition, miR-107 overexpression increased the proliferation and decreased the pyroptosis of chondrocytes induced by LPS and ATP. miR-107 overexpression upregulated the Col II expression while down-regulated the expression of IL-1ß, IL-18, and MMP-13 in supernatant of chondrocytes or chondrocytes induced by LPS and ATP. miR-107 overexpression down-regulated the expression of caspase-1, c-caspase-1, GSDMD-N, and TLR4 in chondrocytes induced by LPS and ATP. Furthermore, miR-107 directly targeted caspase-1. CONCLUSIONS: miR-107 can protect against KOA by downregulating caspase-1 to decrease pyroptosis, thereby promoting collagen protein secreted by chondrocytes by down-regulating IL-1ß.

Data mining techniques for assisting the diagnosis of pressure ulcer development in surgical patients.

Pressure ulcer is a serious problem during patient care processes. The high risk factors in the development of pressure ulcer remain unclear during long surgery. Moreover, past preventive policies are hard to implement in a busy operation room. The objective of this study is to use data mining techniques to construct the prediction model for pressure ulcers. Four data mining techniques, namely, Mahalanobis Taguchi System (MTS), Support Vector Machines (SVMs), decision tree (DT), and logistic regression (LR), are used to select the important attributes from the data to predict the incidence of pressure ulcers. Measurements of sensitivity, specificity, F(1), and g-means were used to compare the performance of four classifiers on the pressure ulcer data set. The results show that data mining techniques obtain good results in predicting the incidence of pressure ulcer. We can conclude that data mining techniques can help identify the important factors and provide a feasible model to predict pressure ulcer development.

Conversion of squid pen by a novel strain Lactobacillus paracasei subsp. paracasei TKU010, and its application in antimicrobial and antioxidants activity.

TKU010 was isolated from infant vomited milk and identified as Lactobacillus paracasei subsp. paracasei. TKU010 had desirable properties concerning its ability to withstand adverse conditions in the gastrointestinal tract. The hydrolysate of casein enhanced the growth of TKU010 most obviously (17.20-18.25 OD(660)), followed by the hydrolysate of SPP (16.00-15.06 OD(660)). Incubating with SPP, both the culture supernatant of TKU010 on the first day and the fourth day showed inhibitory activities on E. coli BCRC13086, F. oxysporum BCRC32121 and A. fumigatus BCRC30099. TKU010 culture supernatant (1% SPP) incubated for 3 days has high antioxidant activity; the DPPH scavenging ability was 75% per ml. Thus, TKU010 could be preferably used as a starter to produce fermented milk with possibly interesting organoleptic properties. Besides, we have shown that squid pen wastes can be utilized to generate a high value-added product, and have revealed its hidden potential in the production of biocontrol agents and functional foods.