Purpose: To develop a machine learning (ML) classifier for predicting post-induction hypotension (PIH) in non-cardiac surgeries. Materials and Methods: Preoperative data and early vital signs were obtained from 3669 cases in the VitalDB database, an opensource registry. PIH was defined as sustained mean arterial pressure (MAP) <65 mm Hg within 20 minutes since induction or from induction to incision. Six different ML algorithms were used to create binary classifiers to predict PIH. The primary outcome was the area under the receiver operating characteristic curve (AUROC) of ML classifiers. Results: A total of 2321 (63.3%) cases exhibited PIH. Among ML classifiers, the random forest regressor and extremely gradient boosting regressor showed the highest AUROC, both recording a value of 0.772. Excluding these models, the light gradient boosting machine regressor showed the second highest AUROC [0.769; 95% confidence interval (CI), 0.767–0.771], followed by the gradient boosting regressor (0.768; 95% CI, 0.763–0.772), AdaBoost regressor (0.752; 95% CI, 0.743–0.761), and automatic relevance determination regression (0.685; 95% CI, 0.669–0.701). The top three important features were mean diastolic blood pressure (DBP), minimum MAP, and minimum DBP from anesthetic induction to tracheal intubation, and these features were lower in cases with PIH (all p<0.001). Conclusion ML classifiers exhibited moderate performance in predicting PIH, and have the potential for real-time prediction.

Purpose: To develop a machine learning (ML) classifier for predicting post-induction hypotension (PIH) in non-cardiac surgeries. Materials and Methods: Preoperative data and early vital signs were obtained from 3669 cases in the VitalDB database, an opensource registry. PIH was defined as sustained mean arterial pressure (MAP) <65 mm Hg within 20 minutes since induction or from induction to incision. Six different ML algorithms were used to create binary classifiers to predict PIH. The primary outcome was the area under the receiver operating characteristic curve (AUROC) of ML classifiers. Results: A total of 2321 (63.3%) cases exhibited PIH. Among ML classifiers, the random forest regressor and extremely gradient boosting regressor showed the highest AUROC, both recording a value of 0.772. Excluding these models, the light gradient boosting machine regressor showed the second highest AUROC [0.769; 95% confidence interval (CI), 0.767–0.771], followed by the gradient boosting regressor (0.768; 95% CI, 0.763–0.772), AdaBoost regressor (0.752; 95% CI, 0.743–0.761), and automatic relevance determination regression (0.685; 95% CI, 0.669–0.701). The top three important features were mean diastolic blood pressure (DBP), minimum MAP, and minimum DBP from anesthetic induction to tracheal intubation, and these features were lower in cases with PIH (all p<0.001). Conclusion ML classifiers exhibited moderate performance in predicting PIH, and have the potential for real-time prediction.

Purpose: To develop a machine learning (ML) classifier for predicting post-induction hypotension (PIH) in non-cardiac surgeries. Materials and Methods: Preoperative data and early vital signs were obtained from 3669 cases in the VitalDB database, an opensource registry. PIH was defined as sustained mean arterial pressure (MAP) <65 mm Hg within 20 minutes since induction or from induction to incision. Six different ML algorithms were used to create binary classifiers to predict PIH. The primary outcome was the area under the receiver operating characteristic curve (AUROC) of ML classifiers. Results: A total of 2321 (63.3%) cases exhibited PIH. Among ML classifiers, the random forest regressor and extremely gradient boosting regressor showed the highest AUROC, both recording a value of 0.772. Excluding these models, the light gradient boosting machine regressor showed the second highest AUROC [0.769; 95% confidence interval (CI), 0.767–0.771], followed by the gradient boosting regressor (0.768; 95% CI, 0.763–0.772), AdaBoost regressor (0.752; 95% CI, 0.743–0.761), and automatic relevance determination regression (0.685; 95% CI, 0.669–0.701). The top three important features were mean diastolic blood pressure (DBP), minimum MAP, and minimum DBP from anesthetic induction to tracheal intubation, and these features were lower in cases with PIH (all p<0.001). Conclusion ML classifiers exhibited moderate performance in predicting PIH, and have the potential for real-time prediction.

Purpose: To develop a machine learning (ML) classifier for predicting post-induction hypotension (PIH) in non-cardiac surgeries. Materials and Methods: Preoperative data and early vital signs were obtained from 3669 cases in the VitalDB database, an opensource registry. PIH was defined as sustained mean arterial pressure (MAP) <65 mm Hg within 20 minutes since induction or from induction to incision. Six different ML algorithms were used to create binary classifiers to predict PIH. The primary outcome was the area under the receiver operating characteristic curve (AUROC) of ML classifiers. Results: A total of 2321 (63.3%) cases exhibited PIH. Among ML classifiers, the random forest regressor and extremely gradient boosting regressor showed the highest AUROC, both recording a value of 0.772. Excluding these models, the light gradient boosting machine regressor showed the second highest AUROC [0.769; 95% confidence interval (CI), 0.767–0.771], followed by the gradient boosting regressor (0.768; 95% CI, 0.763–0.772), AdaBoost regressor (0.752; 95% CI, 0.743–0.761), and automatic relevance determination regression (0.685; 95% CI, 0.669–0.701). The top three important features were mean diastolic blood pressure (DBP), minimum MAP, and minimum DBP from anesthetic induction to tracheal intubation, and these features were lower in cases with PIH (all p<0.001). Conclusion ML classifiers exhibited moderate performance in predicting PIH, and have the potential for real-time prediction.

Purpose: To develop a machine learning (ML) classifier for predicting post-induction hypotension (PIH) in non-cardiac surgeries. Materials and Methods: Preoperative data and early vital signs were obtained from 3669 cases in the VitalDB database, an opensource registry. PIH was defined as sustained mean arterial pressure (MAP) <65 mm Hg within 20 minutes since induction or from induction to incision. Six different ML algorithms were used to create binary classifiers to predict PIH. The primary outcome was the area under the receiver operating characteristic curve (AUROC) of ML classifiers. Results: A total of 2321 (63.3%) cases exhibited PIH. Among ML classifiers, the random forest regressor and extremely gradient boosting regressor showed the highest AUROC, both recording a value of 0.772. Excluding these models, the light gradient boosting machine regressor showed the second highest AUROC [0.769; 95% confidence interval (CI), 0.767–0.771], followed by the gradient boosting regressor (0.768; 95% CI, 0.763–0.772), AdaBoost regressor (0.752; 95% CI, 0.743–0.761), and automatic relevance determination regression (0.685; 95% CI, 0.669–0.701). The top three important features were mean diastolic blood pressure (DBP), minimum MAP, and minimum DBP from anesthetic induction to tracheal intubation, and these features were lower in cases with PIH (all p<0.001). Conclusion ML classifiers exhibited moderate performance in predicting PIH, and have the potential for real-time prediction.

Background and Objectives: Telomerase reverse transcriptase (TERT) promoter mutations are a poor prognostic factor in differentiated thyroid carcinoma (DTC). However, their prognostic value in anaplastic thyroid carcinoma (ATC) is unclear. Therefore, we investigated whether TERT promoter mutations also act as an independent poor prognostic factor in ATC. Materials and Methods: We reviewed the medical records of 41 patients with ATC who underwent the TERT promoter mutations test at Samsung Medical Center between November 1995 and December 2022. The aggressive treatment group was defined as patients who underwent surgery, external radiotherapy, and systemic therapy. Results: Among 41 patients, 15 (36.6%) showed TERT promoter mutations. There only differences in the clinicopathological characteristics between the TERT-mutant and wild-type groups were tumor size and coexistence of DTC. Median tumor size in the TERT-mutant group was 5.1 cm (3.0-11.0), which was significantly larger than that in the wild-type group (4.1 cm, 0.8-8.0, p=0.010). Nevertheless, the TERT-mutant group received relatively more aggressive treatment (53.3% vs. 19.2%, p=0.056), and the overall survival of the TERT-mutant group was longer than that of the wild-type group (9.4 months [0.4-51.5] vs. 7.1 months [0.4-49.5]), but its difference was not significant (p=0.458). In multiple regression analysis, distant metastasis was a significant prognostic factor, but TERT promoter mutation was not. Conclusion Unlike in DTC, TERT promoter mutations were not an independent poor prognostic factor in ATC.

Nasal dermoids are congenital midline nasal lesions that occur along with encephaloceles and gliomas. They can cause both deformity of nasal structure and intracranial infection as they grow. Treatment for these lesions is be concerned with two aspects, the complete removal of the lesions and making the surgical scar cosmetically acceptable. To that goal, many surgical approaches such as vertical incision, transverse incision, lateral rhinotomy and open rhinoplasty have been introduced. A 12-month male child presented with palpable mass at nasal root. The mass was easily movable, non-compressible and did not present fistula. A well-defined cystic mass without intracranial extension was found on the computerized tomography scans. Open rhinoplasty approach was opted for according to the guardians’ preference to avoid visible facial scar, and the lesions were completely resected. The pathologic examination confirmed the lesion to be nasal dermoids. The columellar scar was negligible and there was no recurrence at 5 year-follow up after surgery.

Background: Medullary thyroid carcinoma (MTC) has a poorer prognosis than differentiated thyroid cancers; however, comprehensive data on the long-term outcomes of MTC remain scarce. This study investigated the extended clinical outcomes of MTC and aimed to identify prognostic factors. Methods: Patients diagnosed with MTC between 1980 and 2020 were retrospectively reviewed. Their clinical characteristics, longterm clinical outcomes, and prognostic factors for recurrence and mortality were analyzed. Results: The study included 226 patients (144 women, 82 men). The disease-specific survival (DSS) rates for all MTC patients at 5-, 10-, 20-, and 30-year intervals were 92.7%, 89.4%, 74.3%, and 68.1%, respectively. The recurrence-free survival (RFS) rates were 71.1%, 56.1%, 40.2%, and 32.1% at these intervals. DSS was comparable between the groups from 1980–2009 and 2010–2020 (P=0.995); however, the 1980–2009 group had significantly lower RFS rates (P=0.031). The 2010–2020 group exhibited greater extents of surgical and lymph node dissection (P=0.003) and smaller tumors (P=0.003). Multivariate analysis identified extrathyroidal extension as the strongest prognostic factor for both RFS and DSS. Age >55 years and tumor size of ≥2 cm were also significant prognostic factors for DSS, while hereditary disease and lymph node metastasis were significant for RFS. Survival analysis after propensity-score matching of rearranged during transfection (RET)-negative and non-screened RET-positive groups showed comparable DSS but longer RFS in the RET-negative group. Conclusion Extrathyroidal extension was identified as the strongest prognostic factor for RFS and DSS. Older age and larger tumor size were associated with decreased DSS, while RET mutation and lymph node metastasis significantly impacted RFS.

Nasal dermoids are congenital midline nasal lesions that occur along with encephaloceles and gliomas. They can cause both deformity of nasal structure and intracranial infection as they grow. Treatment for these lesions is be concerned with two aspects, the complete removal of the lesions and making the surgical scar cosmetically acceptable. To that goal, many surgical approaches such as vertical incision, transverse incision, lateral rhinotomy and open rhinoplasty have been introduced. A 12-month male child presented with palpable mass at nasal root. The mass was easily movable, non-compressible and did not present fistula. A well-defined cystic mass without intracranial extension was found on the computerized tomography scans. Open rhinoplasty approach was opted for according to the guardians’ preference to avoid visible facial scar, and the lesions were completely resected. The pathologic examination confirmed the lesion to be nasal dermoids. The columellar scar was negligible and there was no recurrence at 5 year-follow up after surgery.

Nasal dermoids are congenital midline nasal lesions that occur along with encephaloceles and gliomas. They can cause both deformity of nasal structure and intracranial infection as they grow. Treatment for these lesions is be concerned with two aspects, the complete removal of the lesions and making the surgical scar cosmetically acceptable. To that goal, many surgical approaches such as vertical incision, transverse incision, lateral rhinotomy and open rhinoplasty have been introduced. A 12-month male child presented with palpable mass at nasal root. The mass was easily movable, non-compressible and did not present fistula. A well-defined cystic mass without intracranial extension was found on the computerized tomography scans. Open rhinoplasty approach was opted for according to the guardians’ preference to avoid visible facial scar, and the lesions were completely resected. The pathologic examination confirmed the lesion to be nasal dermoids. The columellar scar was negligible and there was no recurrence at 5 year-follow up after surgery.