Purpose: To estimate the sample size required for developing artificial intelligence (AI) that can predict soft-tissue and alveolar bone changes following orthodontic treatment. Materials and Methods: From the original data sets with N=887, consisting of 132 input and 88 output variables used to create AI models for predicting treatment changes following orthodontic treatment, six subsets of the data (n=75, 150, 300, 450, 600, and 750) were generated through random resampling procedures. The process was repeated four times, resulting in 24 different data subsets. Each data subset was used to create a total of 24 AI models using the TabNet deep neural network algorithm. The clinically acceptable prediction accuracy was defined as a less than 1.5 mm prediction error on the lower lip. The prediction errors from each AI model were compared according to sample sizes and analyzed to estimate the optimal sample size. Results: The prediction error decreased with increasing sample sizes. A training sample size greater than approximately 1650 was estimated to develop an AI model with less than 1.5 mm of prediction errors at the lower lip area. Conclusion From a statistical and research design perspective, a considerable amount of training data appears necessary to develop an AI prediction model with clinically acceptable accuracy.

Purpose: To estimate the sample size required for developing artificial intelligence (AI) that can predict soft-tissue and alveolar bone changes following orthodontic treatment. Materials and Methods: From the original data sets with N=887, consisting of 132 input and 88 output variables used to create AI models for predicting treatment changes following orthodontic treatment, six subsets of the data (n=75, 150, 300, 450, 600, and 750) were generated through random resampling procedures. The process was repeated four times, resulting in 24 different data subsets. Each data subset was used to create a total of 24 AI models using the TabNet deep neural network algorithm. The clinically acceptable prediction accuracy was defined as a less than 1.5 mm prediction error on the lower lip. The prediction errors from each AI model were compared according to sample sizes and analyzed to estimate the optimal sample size. Results: The prediction error decreased with increasing sample sizes. A training sample size greater than approximately 1650 was estimated to develop an AI model with less than 1.5 mm of prediction errors at the lower lip area. Conclusion From a statistical and research design perspective, a considerable amount of training data appears necessary to develop an AI prediction model with clinically acceptable accuracy.

Purpose: To estimate the sample size required for developing artificial intelligence (AI) that can predict soft-tissue and alveolar bone changes following orthodontic treatment. Materials and Methods: From the original data sets with N=887, consisting of 132 input and 88 output variables used to create AI models for predicting treatment changes following orthodontic treatment, six subsets of the data (n=75, 150, 300, 450, 600, and 750) were generated through random resampling procedures. The process was repeated four times, resulting in 24 different data subsets. Each data subset was used to create a total of 24 AI models using the TabNet deep neural network algorithm. The clinically acceptable prediction accuracy was defined as a less than 1.5 mm prediction error on the lower lip. The prediction errors from each AI model were compared according to sample sizes and analyzed to estimate the optimal sample size. Results: The prediction error decreased with increasing sample sizes. A training sample size greater than approximately 1650 was estimated to develop an AI model with less than 1.5 mm of prediction errors at the lower lip area. Conclusion From a statistical and research design perspective, a considerable amount of training data appears necessary to develop an AI prediction model with clinically acceptable accuracy.

Purpose: To estimate the sample size required for developing artificial intelligence (AI) that can predict soft-tissue and alveolar bone changes following orthodontic treatment. Materials and Methods: From the original data sets with N=887, consisting of 132 input and 88 output variables used to create AI models for predicting treatment changes following orthodontic treatment, six subsets of the data (n=75, 150, 300, 450, 600, and 750) were generated through random resampling procedures. The process was repeated four times, resulting in 24 different data subsets. Each data subset was used to create a total of 24 AI models using the TabNet deep neural network algorithm. The clinically acceptable prediction accuracy was defined as a less than 1.5 mm prediction error on the lower lip. The prediction errors from each AI model were compared according to sample sizes and analyzed to estimate the optimal sample size. Results: The prediction error decreased with increasing sample sizes. A training sample size greater than approximately 1650 was estimated to develop an AI model with less than 1.5 mm of prediction errors at the lower lip area. Conclusion From a statistical and research design perspective, a considerable amount of training data appears necessary to develop an AI prediction model with clinically acceptable accuracy.

KCNQ2 mutations induce a neonatal-onset epileptic encephalopathy of widely varying severity, ranging from benign familial neonatal epilepsy to severe refractory epileptic encephalopathy. Refractory seizures with KCNQ2 mutations have a positive response to sodium-channel blockers. Recently, a distinctive ictal pattern has been reported during amplitude-integrated electroencephalographic (aEEG) monitoring in infants with KCNQ2 encephalopathy. Herein, we describe a case of KCNQ2 encephalopathy with this distinctive ictal aEEG pattern, which was confirmed using conventional electroencephalography (EEG). A 3-day-old female infant presented with neonatal seizures accompanied by cyanosis and desaturation. Her seizure semiology was tonic and focal clonic. Her ictal aEEG demonstrated a sudden rise in amplitude followed by a suppressed background pattern. This pattern was also confirmed on conventional EEG. Her seizures were refractory despite the administration of multiple conventional antiepileptic drugs. Finally, c.794C>T; p. (Ala265Val) mutation was observed in the KCNQ2 gene on genetic testing, and she was diagnosed with KCNQ2 encephalopathy. Identifying this distinctive ictal pattern on aEEG monitoring facilitates the early detection of KCNQ2 encephalopathy and timely targeted treatment in patients with refractory seizures.

KCNQ2 mutations induce a neonatal-onset epileptic encephalopathy of widely varying severity, ranging from benign familial neonatal epilepsy to severe refractory epileptic encephalopathy. Refractory seizures with KCNQ2 mutations have a positive response to sodium-channel blockers. Recently, a distinctive ictal pattern has been reported during amplitude-integrated electroencephalographic (aEEG) monitoring in infants with KCNQ2 encephalopathy. Herein, we describe a case of KCNQ2 encephalopathy with this distinctive ictal aEEG pattern, which was confirmed using conventional electroencephalography (EEG). A 3-day-old female infant presented with neonatal seizures accompanied by cyanosis and desaturation. Her seizure semiology was tonic and focal clonic. Her ictal aEEG demonstrated a sudden rise in amplitude followed by a suppressed background pattern. This pattern was also confirmed on conventional EEG. Her seizures were refractory despite the administration of multiple conventional antiepileptic drugs. Finally, c.794C>T; p. (Ala265Val) mutation was observed in the KCNQ2 gene on genetic testing, and she was diagnosed with KCNQ2 encephalopathy. Identifying this distinctive ictal pattern on aEEG monitoring facilitates the early detection of KCNQ2 encephalopathy and timely targeted treatment in patients with refractory seizures.

BACKGROUND/OBJECTIVES: Okra seed is a rich source of various nutritional and bioactive constituents, but its mechanism of action is still unclear. The aim of this study was to evaluated the effects on glucose uptake and serum lipid profiles of unsaponifiable matter (USM) from okra seed in adipocytes and diabetic animal models.MATERIALS/METHODSUSM was prepared from okra seed powder by saponification. The contents of phytosterols and vitamin E in USM were measured. 3T3-L1 preadipocytes were cultured for 6 days with different concentrations of USM (0–200 μg/mL). The diabetic rats were administered with or without USM for 5 wk. RESULTS: In the USM, the contents of phytosterols and vitamin E were 394.13 mg/g USM and 31.16 mg/g USM, respectively. USM showed no cytotoxicity and led to an approximately 1.4-fold increase in glucose uptake in 3T3-L1 adipocytes. The treatment of USM also increased the expressions of peroxisome proliferator-activated receptor-γ and glucose transporter-4 in a dose-dependent manner in adipocytes. The body weight change was not significantly different in all diabetic rats. However, blood glucose and the weights of liver and adipose tissues were significantly reduced compared to those in the control diabetic rats. Treatment with USM decreased the levels of triglycerides, total cholesterol, and low-density lipoprotein cholesterol compared to the control group. The USM group also showed significantly decreased atherogenic indices and cardiac risk factors. CONCLUSION These results suggest that USM from okra seed improves the hypoglycemic and hypolipidemic effects in diabetic rats, and provides valuable information for improving the functional properties of okra seed.

Topoisomerase IIα has been a representative anti-cancer target for decades thanks to its functional necessity in highly proliferative cancer cells. As type of topoisomerase IIα targeting drugs, topoisomerase II poisons are frequently in clinical usage. However, topoisomerase II poisons result in crucial consequences resulted from mechanistically induced DNA toxicity. For this reason, it is needed to develop catalytic inhibitors of topoisomerase IIα through the alternative mechanism of enzymatic regulation. As a catalytic inhibitor of topoisomerase IIα, AK-I-191 was previously reported for its enzyme inhibitory activity. In this study, we clarified the mechanism of AK-I-191 and conducted various types of spectroscopic and biological evaluations for deeper understanding of its mechanism of action. Conclusively, AK-I-191 represented potent topoisomerase IIα inhibitory activity through binding to minor groove of DNA double helix and showed synergistic effects with tamoxifen in antiproliferative activity.

Topoisomerase IIα has been a representative anti-cancer target for decades thanks to its functional necessity in highly proliferative cancer cells. As type of topoisomerase IIα targeting drugs, topoisomerase II poisons are frequently in clinical usage. However, topoisomerase II poisons result in crucial consequences resulted from mechanistically induced DNA toxicity. For this reason, it is needed to develop catalytic inhibitors of topoisomerase IIα through the alternative mechanism of enzymatic regulation. As a catalytic inhibitor of topoisomerase IIα, AK-I-191 was previously reported for its enzyme inhibitory activity. In this study, we clarified the mechanism of AK-I-191 and conducted various types of spectroscopic and biological evaluations for deeper understanding of its mechanism of action. Conclusively, AK-I-191 represented potent topoisomerase IIα inhibitory activity through binding to minor groove of DNA double helix and showed synergistic effects with tamoxifen in antiproliferative activity.