Purpose: This study explores the potential of artificial intelligence (AI) in optimizing radiotherapy protocols for personalized cancer treatment. Specifically, it investigates the role of AI-based segmentation tools in improving accuracy and efficiency across various anatomical regions. Methods: A dataset of 500 anonymized patient computed tomography scans from Chungbuk National University Hospital was used to develop and validate AI models for segmenting organs-atrisk. The models were tailored for five anatomical regions: head and neck, chest, abdomen, breast, and pelvis. Performance was evaluated using Dice Similarity Coefficient (DSC), Mean Surface Distance, and the 95th Percentile Hausdorff Distance (HD95). Results: The AI models achieved high segmentation accuracy for large, well-defined structures such as the brain, lungs, and liver, with DSC values exceeding 0.95 in many cases. However, challenges were observed for smaller or complex structures, including the optic chiasm and rectum, with instances of segmentation failure and infinity values for HD95. These findings highlight the variability in performance depending on anatomical complexity and structure size. Conclusions AI-based segmentation tools demonstrate significant potential to streamline radiotherapy workflows, reduce inter-observer variability, and enhance treatment accuracy. Despite challenges with smaller structures, the integration of AI enables dynamic, patient-specific adaptations to anatomical changes, contributing to more precise and effective cancer treatments.Future work should focus on refining models for anatomically complex structures and validating these methods in diverse clinical settings.

Purpose: This study explores the potential of artificial intelligence (AI) in optimizing radiotherapy protocols for personalized cancer treatment. Specifically, it investigates the role of AI-based segmentation tools in improving accuracy and efficiency across various anatomical regions. Methods: A dataset of 500 anonymized patient computed tomography scans from Chungbuk National University Hospital was used to develop and validate AI models for segmenting organs-atrisk. The models were tailored for five anatomical regions: head and neck, chest, abdomen, breast, and pelvis. Performance was evaluated using Dice Similarity Coefficient (DSC), Mean Surface Distance, and the 95th Percentile Hausdorff Distance (HD95). Results: The AI models achieved high segmentation accuracy for large, well-defined structures such as the brain, lungs, and liver, with DSC values exceeding 0.95 in many cases. However, challenges were observed for smaller or complex structures, including the optic chiasm and rectum, with instances of segmentation failure and infinity values for HD95. These findings highlight the variability in performance depending on anatomical complexity and structure size. Conclusions AI-based segmentation tools demonstrate significant potential to streamline radiotherapy workflows, reduce inter-observer variability, and enhance treatment accuracy. Despite challenges with smaller structures, the integration of AI enables dynamic, patient-specific adaptations to anatomical changes, contributing to more precise and effective cancer treatments.Future work should focus on refining models for anatomically complex structures and validating these methods in diverse clinical settings.

Purpose: This study explores the potential of artificial intelligence (AI) in optimizing radiotherapy protocols for personalized cancer treatment. Specifically, it investigates the role of AI-based segmentation tools in improving accuracy and efficiency across various anatomical regions. Methods: A dataset of 500 anonymized patient computed tomography scans from Chungbuk National University Hospital was used to develop and validate AI models for segmenting organs-atrisk. The models were tailored for five anatomical regions: head and neck, chest, abdomen, breast, and pelvis. Performance was evaluated using Dice Similarity Coefficient (DSC), Mean Surface Distance, and the 95th Percentile Hausdorff Distance (HD95). Results: The AI models achieved high segmentation accuracy for large, well-defined structures such as the brain, lungs, and liver, with DSC values exceeding 0.95 in many cases. However, challenges were observed for smaller or complex structures, including the optic chiasm and rectum, with instances of segmentation failure and infinity values for HD95. These findings highlight the variability in performance depending on anatomical complexity and structure size. Conclusions AI-based segmentation tools demonstrate significant potential to streamline radiotherapy workflows, reduce inter-observer variability, and enhance treatment accuracy. Despite challenges with smaller structures, the integration of AI enables dynamic, patient-specific adaptations to anatomical changes, contributing to more precise and effective cancer treatments.Future work should focus on refining models for anatomically complex structures and validating these methods in diverse clinical settings.

Purpose: This study explores the potential of artificial intelligence (AI) in optimizing radiotherapy protocols for personalized cancer treatment. Specifically, it investigates the role of AI-based segmentation tools in improving accuracy and efficiency across various anatomical regions. Methods: A dataset of 500 anonymized patient computed tomography scans from Chungbuk National University Hospital was used to develop and validate AI models for segmenting organs-atrisk. The models were tailored for five anatomical regions: head and neck, chest, abdomen, breast, and pelvis. Performance was evaluated using Dice Similarity Coefficient (DSC), Mean Surface Distance, and the 95th Percentile Hausdorff Distance (HD95). Results: The AI models achieved high segmentation accuracy for large, well-defined structures such as the brain, lungs, and liver, with DSC values exceeding 0.95 in many cases. However, challenges were observed for smaller or complex structures, including the optic chiasm and rectum, with instances of segmentation failure and infinity values for HD95. These findings highlight the variability in performance depending on anatomical complexity and structure size. Conclusions AI-based segmentation tools demonstrate significant potential to streamline radiotherapy workflows, reduce inter-observer variability, and enhance treatment accuracy. Despite challenges with smaller structures, the integration of AI enables dynamic, patient-specific adaptations to anatomical changes, contributing to more precise and effective cancer treatments.Future work should focus on refining models for anatomically complex structures and validating these methods in diverse clinical settings.

Purpose: This study explores the potential of artificial intelligence (AI) in optimizing radiotherapy protocols for personalized cancer treatment. Specifically, it investigates the role of AI-based segmentation tools in improving accuracy and efficiency across various anatomical regions. Methods: A dataset of 500 anonymized patient computed tomography scans from Chungbuk National University Hospital was used to develop and validate AI models for segmenting organs-atrisk. The models were tailored for five anatomical regions: head and neck, chest, abdomen, breast, and pelvis. Performance was evaluated using Dice Similarity Coefficient (DSC), Mean Surface Distance, and the 95th Percentile Hausdorff Distance (HD95). Results: The AI models achieved high segmentation accuracy for large, well-defined structures such as the brain, lungs, and liver, with DSC values exceeding 0.95 in many cases. However, challenges were observed for smaller or complex structures, including the optic chiasm and rectum, with instances of segmentation failure and infinity values for HD95. These findings highlight the variability in performance depending on anatomical complexity and structure size. Conclusions AI-based segmentation tools demonstrate significant potential to streamline radiotherapy workflows, reduce inter-observer variability, and enhance treatment accuracy. Despite challenges with smaller structures, the integration of AI enables dynamic, patient-specific adaptations to anatomical changes, contributing to more precise and effective cancer treatments.Future work should focus on refining models for anatomically complex structures and validating these methods in diverse clinical settings.

Purpose: We aimed to investigate the risk factors and patterns of locoregional recurrence (LRR) after radical nephrectomy (RN) in patients with locally advanced renal cell carcinoma (RCC). Materials and Methods: We retrospectively analyzed 245 patients who underwent RN for non-metastatic pT3-4 RCC from January 2006 to January 2016. We analyzed the risk factors associated with poor locoregional control using Cox regression. Anatomical mapping was performed on reference computed tomography scans showing intact kidneys. Results: The median follow-up duration was 56 months (range, 1 to 128 months). Tumor extension to renal vessels or the inferior vena cava (IVC) and Fuhrman’s nuclear grade IV were identified as independent risk factors of LRR. The 5-year actuarial LRR rates in groups with no risk factor, one risk factor, and two risk factors were 2.3%, 19.8%, and 30.8%, respectively (p < 0.001). The locations of LRR were distributed as follows: aortocaval area (n=2), paraaortic area (n=4), retrocaval area (n=5), and tumor bed (n=11). No LRR was observed above the celiac axis (CA) or under the inferior mesenteric artery (IMA). Conclusion Tumor extension to renal vessels or the IVC and Fuhrman’s nuclear grade IV were the independent risk factors associated with LRR after RN for pT3-4 RCC. The locations of LRR after RN for RCC were distributed in the tumor bed and regional lymphatic area from the bifurcation of the CA to that of the IMA.

Background: and Purpose We aimed to develop a model predicting early recanalization after intravenous tissue plasminogen activator (t-PA) treatment in large-vessel occlusion. Methods: Using data from two different multicenter prospective cohorts, we determined the factors associated with early recanalization immediately after t-PA in stroke patients with large-vessel occlusion, and developed and validated a prediction model for early recanalization. Clot volume was semiautomatically measured on thin-section computed tomography using software, and the degree of collaterals was determined using the Tan score. Follow-up angiographic studies were performed immediately after t-PA treatment to assess early recanalization. Results: Early recanalization, assessed 61.0±44.7 minutes after t-PA bolus, was achieved in 15.5% (15/97) in the derivation cohort and in 10.5% (8/76) in the validation cohort. Clot volume (odds ratio [OR], 0.979; 95% confidence interval [CI], 0.961 to 0.997; P=0.020) and good collaterals (OR, 6.129; 95% CI, 1.592 to 23.594; P=0.008) were significant factors associated with early recanalization. The area under the curve (AUC) of the model including clot volume was 0.819 (95% CI, 0.720 to 0.917) and 0.842 (95% CI, 0.746 to 0.938) in the derivation and validation cohorts, respectively. The AUC improved when good collaterals were added (derivation cohort: AUC, 0.876; 95% CI, 0.802 to 0.950; P=0.164; validation cohort: AUC, 0.949; 95% CI, 0.886 to 1.000; P=0.036). The integrated discrimination improvement also showed significantly improved prediction (0.097; 95% CI, 0.009 to 0.185; P=0.032). Conclusions The model using clot volume and collaterals predicted early recanalization after intravenous t-PA and had a high performance. This model may aid in determining the recanalization treatment strategy in stroke patients with large-vessel occlusion.

Background: and Purpose We aimed to develop a model predicting early recanalization after intravenous tissue plasminogen activator (t-PA) treatment in large-vessel occlusion. Methods: Using data from two different multicenter prospective cohorts, we determined the factors associated with early recanalization immediately after t-PA in stroke patients with large-vessel occlusion, and developed and validated a prediction model for early recanalization. Clot volume was semiautomatically measured on thin-section computed tomography using software, and the degree of collaterals was determined using the Tan score. Follow-up angiographic studies were performed immediately after t-PA treatment to assess early recanalization. Results: Early recanalization, assessed 61.0±44.7 minutes after t-PA bolus, was achieved in 15.5% (15/97) in the derivation cohort and in 10.5% (8/76) in the validation cohort. Clot volume (odds ratio [OR], 0.979; 95% confidence interval [CI], 0.961 to 0.997; P=0.020) and good collaterals (OR, 6.129; 95% CI, 1.592 to 23.594; P=0.008) were significant factors associated with early recanalization. The area under the curve (AUC) of the model including clot volume was 0.819 (95% CI, 0.720 to 0.917) and 0.842 (95% CI, 0.746 to 0.938) in the derivation and validation cohorts, respectively. The AUC improved when good collaterals were added (derivation cohort: AUC, 0.876; 95% CI, 0.802 to 0.950; P=0.164; validation cohort: AUC, 0.949; 95% CI, 0.886 to 1.000; P=0.036). The integrated discrimination improvement also showed significantly improved prediction (0.097; 95% CI, 0.009 to 0.185; P=0.032). Conclusions The model using clot volume and collaterals predicted early recanalization after intravenous t-PA and had a high performance. This model may aid in determining the recanalization treatment strategy in stroke patients with large-vessel occlusion.

Purpose: For unresectable or initially metastatic gastric cancer, conversion surgery (CVS), after systemic chemotherapy, has received attention as a treatment strategy. This study evaluated the prognostic value of ypTNM stage and the oncologic outcomes in patients receiving CVS. Materials and Methods: A retrospective review of clinicopathologic findings and oncologic outcomes of 116 patients who underwent CVS with curative intent, after combination chemotherapy, between January 2000 and December 2015, has been reported here. Results: Twenty-six patients (22.4%) underwent combined resection of another organ and 12 patients received para-aortic lymphadenectomy (10.3%). Pathologic complete remission (CR) was confirmed in 11 cases (9.5%). The median overall survival (OS) and disease-free survival (DFS) times were 35.0 and 21.3 months, respectively. In multivariate analysis, ypTNM stage was the sole independent prognostic factor for DFS (P=0.042). Tumors invading an adjacent organ or involving distant lymph nodes showed better survival than those with peritoneal seeding or solid organ metastasis (P=0.084). Kaplan-Meier curves showed that the 3-year OS rate of patients with pathologic CR and those with CR of the primary tumor but residual node metastasis was 81.8% and 80.0%, respectively. OS was 65.8% for stage 1 patients, 49.8% for those at stage 2, and 36.3% for those at stage 3. Conclusions The ypTNM staging is a significant prognostic factor in patients who underwent CVS for localized unresectable or stage IV gastric cancers. Patients with locally advanced but unresectable lesions or with tumors with distant nodal metastasis may be good candidates for CVS.

BACKGROUND/AIMS: This study aimed to compare the clinical characteristics and management patterns of inflammatory bowel disease (IBD) patients in a secondary hospital (SH) with those in tertiary referral centers (TRC). METHODS: Data from IBD patients in SH and 2 TRCs were retrospectively reviewed. The cumulative thiopurine use rate was compared between hospitals after controlling for different baseline characteristics using propensity score matching. RESULTS: Among the total of 447 patients with IBD, 178 Crohn's disease (CD) and 269 ulcerative colitis (UC) patients were included. Regarding initial CD symptoms, patients from SH were more likely to show perianal symptoms, such as anal pain or discharge (56.6% vs. 34.3%, P=0.003), whereas those from TRCs more often had luminal symptoms, such as abdominal pain (54.9% vs. 17.1%, P < 0.001), diarrhea (44.1% vs. 18.4%, P < 0.001), and body weight loss (9.8% vs. 1.3%, P=0.025). Complicating behaviors, such as stricturing and penetrating, were significantly higher in TRCs, while perianal disease was more common in SH. Ileal location was more frequently observed in TRCs. For UC, SH had a more limited extent of disease (proctitis 58.8% vs. 21.2%, P < 0.001). The cumulative azathioprine use rate in SH was significantly lower than that in TRCs in both CD and UC patients after controlling for disease behavior, location, and perianal disease of CD and extent of UC. CONCLUSIONS: The clinical characteristics and management of the IBD patients in SH were substantially different from those in TRCs. Thiopurine treatment was less commonly used for SH patients.
Abdominal Pain ; Azathioprine ; Body Weight ; Colitis, Ulcerative ; Crohn Disease ; Diarrhea ; Humans ; Inflammatory Bowel Diseases* ; Phenobarbital ; Propensity Score* ; Retrospective Studies ; Secondary Care ; Tertiary Care Centers*