Accurate prediction of a patient's length of stay in the hospital is important to enable efficient and effective management of hospital beds. This paper studies length-of-stay prediction for pediatric patients with respiratory diseases using three decision tree methods: bagging, Adaboost, and random forest. A data set of 11,206 records retrieved from the hospital information system is used for analysis after preprocessing and transformation into a numerical data type using a computation method and an expansion method. Two testing methods, a bisection test and a periodic test, are designed to assess performance. Bagging shows the best results on the bisection test (0.296 RMSE, 0.831 ${R^{2}}$ , and ${0.723 \pe 1}$ ) on the testing set of the whole data test. The performances of the three methods are similar on the periodic test, whereas Adaboost performs slightly better than the other two methods. This study also investigates the importance of various data fields to the prediction of length of stay. The result of the study indicates that decision tree methods can be used for LOS prediction.

Seahorses are increasingly recognized for their nutritional potential, which underscores the necessity for comprehensive biochemical analyses. This study aims to investigate the fatty acid and amino acid compositions of eight seahorse species, including both genders of Hippocampus trimaculatus, Hippocampus kelloggi, Hippocampus abdominalis, and Hippocampus erectus, to evaluate their nutritional value. We employed Gas Chromatography-Mass Spectrometry (GC-MS) and High-Performance Liquid Chromatography (HPLC) to analyze the fatty acid and amino acid profiles of the seahorse species. GC-MS was used to detect 34 fatty acid methyl esters, while HPLC provided detailed amino acid profiles. GC-MS analysis demonstrated high precision with relative standard deviations (RSDs) generally below 2.53 %, satisfactory repeatability (RSDs from 6.55 % to 8.73 %), and stability (RSDs below 2.82 %). Recovery rates for major fatty acids ranged from 98.73 % to 109.12 %. HPLC analysis showed strong separation of amino acid profiles with theoretical plate numbers exceeding 5000. Precision tests yielded RSDs below 1.23 %, with reproducibility and stability tests showing RSDs below 2.73 % and 2.86 %, respectively. Amino acid recovery rates ranged from 97.58 % to 104.66 %. Nutritional analysis revealed significant variations in fatty acid content among the species. Female H. erectus showed higher levels of hexadecanoic acid and saturated fatty acids, while male H. abdominalis had lower concentrations of n-3 full cis 4,7,10,13,16,19-docosahexaenoic acid (DHA). Total lipid yields varied from 3.2491 % to 12.3175 %, with major fatty acids constituting 17.9717 %-74.6962 % of total lipids. In conclusion, this study provides essential insights into the fatty acid and amino acid composition of seahorses, supporting their potential as valuable dietary supplements. The differences between genders in specific fatty acids suggest a nuanced nutritional profile that could be exploited for targeted dietary applications. Further research is needed to explore the seasonal and environmental variations affecting seahorse biochemical composition.