To determine whether radiotherapy ( RT ) can increase pelvic fracture risk in rectal cancer survivors. Rectal cancer patients who underwent curative surgery between 1996 and 2011 in Taiwan were retrospectively studied using the National Health Insurance Research Database ( NHIRD ) of Taiwan. ICD ‐9 Codes 808, 805.4‐805.7, 806.4‐806.7, and 820 (including pelvic, sacrum, lumbar, and femoral neck fracture) were defined as pelvic fracture. Propensity scores for RT , age, and sex were used to perform one‐to‐one matches between the RT and non‐ RT group. Risks of pelvic and arm fractures were compared by multivariable Cox regression. Of the 32 689 patients, 7807 (23.9%) received RT , and 1616 suffered from a pelvic fracture (incidence rate: 1.17/100 person‐years). The median time to pelvic fracture was 2.47 years. After matching, 6952 patients each in the RT and non‐ RT groups were analyzed. RT was associated with an increased risk of pelvic fractures in the multivariable Cox model (hazard ratio ( HR ): 1.246, 95% confidence interval ( CI ): 1.037‐1.495, P = 0.019) but not with arm fractures ( HR : 1.013, 95% CI : 0.814‐1.259, P = 0.911). Subgroup analyses revealed that RT was associated with a higher pelvic fracture rate in women ( HR : 1.431, 95% CI : 1.117‐1.834) but not in men, and the interaction between sex and RT was significant ( P = 0.03). The HR of pelvic fracture increased 2‐4 years after RT ( HR : 1.707, 95% CI : 1.150‐2.534, P = 0.008). An increased risk of pelvic fracture is noted in rectal cancer survivors, especially women, who receive RT .

PurposeSelective internal radiation therapy (SIRT) is an effective treatment strategy for unresectable hepatocellular carcinoma (HCC) patients. However, the prognoses of patients with portal vein thrombosis, extra-hepatic metastases, or residual tumors remain poor when treated with SIRT alone. In these patients, sequential external beam radiotherapy (EBRT) may offer a chance of salvage. Here, we reported the clinical outcomes and the detailed dosimetry analysis of 22 patients treated with combination therapy.MethodsBetween October 2011 and May 2015, 22 consecutive patients who underwent EBRT after yttrium-90 (90Y) SIRT were included in this study. The post-SIRT 90Y bremsstrahlung SPECT/CT of each patient was transferred to dose distribution by adopting the local deposition hypothesis. The patient-specific 3-dimensional biological effective dose distribution of combined SIRT and EBRT was generated. The overall survival and safety were evaluated. The relationship between dosimetric parameters and liver toxicity was analyzed.ResultsThe mean administered activity of SIRT was 1.50 GBq (range: 0.5–2.8). The mean prescribed dose of EBRT was 42.3 Gy (range: 15–63) in 14 fractions (range: 5–15) and was targeted to the residual liver tumor in 12 patients (55%), portal vein thrombosis in 11 patients (50%), and perihilar lymphadenopathies in 4 patients (18%). The overall 1-, 2-, and 3-year survival rates were 59.8%, 47.9%, and 47.9%, respectively. Overall, 8 patients (36%) developed > grade 2 liver toxicities, and the Child-Pugh score prior to EBRT strongly affected the toxicity risk. A dosimetry analysis restricted to 18 Child-Pugh A/B patients showed that the V100 (The fraction of normal liver exposed to more than 100 Gy) to V140 significance differed between patients who did or did not experience hepatotoxicity. The V110 was the strongest predictor of hepatotoxicity (18.6±11.6% vs 29.5±5.8%; P = 0.030).ConclusionCombined therapy is feasible and safe if patients are carefully selected. Specifically, 3-dimensional dosimetry is crucial for the evaluation of efficacy and toxicity. The normal liver V100 to V140 values of the combined dose should be as low as possible to minimize the risk of liver toxicity.

Aim To develop an artificial intelligence-based approach with multi-labeling capability to identify both ST-elevation myocardial infarction (STEMI) and 12 heart rhythms based on 12-lead ECGs. Methods We trained, validated, and tested a long short-term memory (LSTM) model for the multi-label diagnosis of 13 ECG patterns (STEMI+12 rhythm classes) using 60,537 clinical ECGs from 35,981 patients recorded between Jan 15, 2009 and Dec 31, 2018. In addition to the internal test above, we conducted a real-world external test, comparing the LSTM model with board-certified physicians of different specialties using a separate dataset of 308 ECGs covering all 13 ECG diagnoses. Results In the internal test, the area under curves (AUCs) of the LSTM model in classifying the 13 ECG patterns ranged between 0.939 and 0.999. For the external test, the LSTM model for multi-labeling of the 13 ECG patterns evaluated by AUC was 0.987±0.021, which was superior to those of cardiologists (0.898±0.113, P < 0.001), emergency physicians (0.820±0.134, P < 0.001), internists (0.765±0.155, P < 0.001), and a commercial algorithm (0.845±0.121, P < 0.001). Of note, the LSTM model achieved an accuracy of 0.987, AUC of 0.997, and precision, recall, and F  1 score of 0.952, 0.870, and 0.909, respectively, in detecting STEMI. Conclusions We demonstrated the usefulness of an LSTM model in the multi-labeling detection of both rhythm classes and STEMI in competitive testing against board-certified physicians. This AI tool exceeding the cardiologist-level performance in detecting STEMI and rhythm classes on 12-lead ECG may be useful in prioritizing chest pain triage and expediting clinical decision making in healthcare.