BACKGROUND: Over 20 million colonoscopies are performed in China annually. An automatic clinical decision support system (CDSS) with accurate semantic recognition of colonoscopy reports and guideline-based is helpful to relieve the increasing medical burden and standardize the healthcare. In this study, the CDSS was built under a hierarchical-label interpretable classification framework, trained by a state-of-the-art transformer-based model, and validated in a multi-center style. METHODS: We conducted stratified sampling on a previously established dataset containing 302,965 electronic colonoscopy reports with pathology, identified 2041 patients' records representative of overall features, and randomly divided into the training and testing sets (7:3). A total of five main labels and 22 sublabels were applied to annotate each record on a network platform, and the data were trained respectively by three pre-training models on Chinese corpus website, including bidirectional encoder representations from transformers (BERT)-base-Chinese (BC), the BERT-wwm-ext-Chinese (BWEC), and ernie-3.0-base-zh (E3BZ). The performance of trained models was subsequently compared with a randomly initialized model, and the preferred model was selected. Model fine-tuning was applied to further enhance the capacity. The system was validated in five other hospitals with 3177 consecutive colonoscopy cases. RESULTS: The E3BZ pre-trained model exhibited the best performance, with a 90.18% accuracy and a 69.14% Macro-F1 score overall. The model achieved 100% accuracy in identifying cancer cases and 99.16% for normal cases. In external validation, the model exhibited favorable consistency and good performance among five hospitals. CONCLUSIONS The novel CDSS possesses high-level semantic recognition of colonoscopy reports, provides appropriate recommendations, and holds the potential to be a powerful tool for physicians and patients. The hierarchical multi-label strategy and pre-training method should be amendable to manage more medical text in the future.
Humans ; Colonoscopy/methods* ; Deep Learning ; Decision Support Systems, Clinical ; Female ; Male

Objective:To explore the dosimetric characteristics of intensity modulated proton therapy (IMPT), intensity modulated radiation therapy (IMRT) and tomotherapy (TOMO) techniques applied in the irradiation of pediatric whole central nervous system tumors.Methods:Taking the target area of a 14-year-old pediatric patient clinically diagnosed with atypical teratoid/rhabdomyoid tumor, meningeal metastasis by Shandong Cancer Hospital and Institute, and undergoing craniospinal irradiation (CSI) as an example, IMPT, IMRT and TOMO plans were designed respectively based on the clinical prescription of the target area and the limit requirements of organs at risk (OARs). The conformal index (CI), homogeneity index (HI) and gradient index (GI) of each planning target volume, as well as the dose volume index of normal tissues, were evaluated to compare the dosimetric characteristics of the three types of plans.Results:The CI (0.71), HI (0.05) and GI (3.13) of the IMPT plan were comparable to those of IMRT plan (0.80, 0.08, 3.14). The HI (0.03) and GI (2.54) of the TOMO plan were excellent, which were all within the clinically acceptable range. The irradiation dose to parallel organs in the IMPT plan was lower than that in the IMRT and TOMO plan. In the IMPT plan, V 5 of lungs was 2.9%, IMRT plan was 37.6%, and TOMO plan was 43.5%. The D mean of liver in the IMPT plan was 0.01 Gy (RBE), IMRT plan was 6.12 Gy, and TOMO plan was 6.39 Gy. In the IMPT plan, none of the bladder, rectum, and femoral head received the dose, while there was low-dose radiation in both IMRT and TOMO plan. For serial organs adjacent to and within the target area, the D max of spinal cord and brainstem in IMPT plan was 39.89 and 39.88 Gy (RBE), respectively; in IMRT plan, they were 39.43 and 38.59 Gy, respectively; and in TOMO plan, they were 38.41 and 37.69 Gy, respectively. The low-dose area in the IMPT plan was significantly better than the photon radiotherapy plans. Among them, the absolute volume IMPT plan occupied by 10% of the prescribed dose area in the patient's body was reduced by 70.10% compared with IMRT plan and 76.96% compared with TOMO plan; the 30% prescribed dose volume IMPT plan was reduced by 53.49% compared with IMRT plan and 62.51% compared with TOMO plan; the 50% prescribed dose volume IMPT plan was reduced by 39.06% compared with IMRT plan and 42.23% compared with TOMO plan. Conclusions:The IMPT plan demonstrated significantly reduced low-dose exposure and lower doses to parallel OARs compared to both IMRT and TOMO plans in pediatric CSI. The CI, HI and GI of the three plans can all meet the clinical requirements. However, for serial organs adjacent to and within the target area, the D max of the IMPT plan may be higher than that of IMRT and TOMO plans.

Objective:To explore the dosimetric characteristics of proton and photon radiotherapy in the treatment of breast cancer.Methods:Four female breast cancer patients who needed radiotherapy at Shandong Cancer Hospital and Institute from January 2024 to May 2024 were selected as the research subjects. The target area ranges of 4 patients were left-sided breast cancer with lymph node involvement, left-sided breast cancer with lymph node involvement and internal mammary node, right-sided breast cancer with lymph node involvement and internal mammary node and synchronous bilateral breast cancer. Intensity modulated proton therapy (IMPT) and intensity modulated radiation therapy (IMRT) plans were designed respectively based on the prescribed dose in the target area and the limits of organs at risk (tomotherapy plan for bilateral breasts). The conformity index (CI), homogeneity index (HI), gradient index (GI) and organs at risk doses were evaluated. The dosimetric characteristics of IMPT and photon radiotherapy were compared.Results:Both IMPT and photon radiotherapy plans of the 4 breast cancer cases met the clinical dose requirements. The HI value of IMPT plans (0.10-0.14) was comparable to that of photon radiotherapy plans (0.10-0.12), and the average CI of the photon radiotherapy plans was 0.10 higher than that of the IMPT plans, and the average GI was 0.55 lower than that of the IMPT plans. The D mean of ipsilateral lung and heart of IMPT was lower, especially in the low-dose area (V 0-3), which was significantly lower than the photon radiotherapy plans, D mean of ipsilateral lung was reduced by 12.2%, 6.1%, 16.1% and 34.8%, respectively, D mean of heart was reduced by 47.2%, 57.0%, 72.4% and 83.0%, respectively. The ipsilateral lung V 20 of IMPT was not lower than photon radiotherapy plans (unilateral breast: IMPT was 30.0%-34.0%, IMRT was 29.0%-35.9%) . Conclusions:IMPT significantly reduces the D mean to the ipsilateral lung and heart while ensuring dose coverage of the target in breast cancer, preventing more volume of surrounding normal tissues from being irradiated. However, IMPT does not show much more advantage than photon radiotherapy plans in the ipsilateral lung V 20.

Objective:To explore the dosimetric characteristics of intensity modulated proton therapy (IMPT) and intensity modulated radiation therapy (IMRT) for typical abdominal and pelvic tumors.Methods:Three patients with abdominal and pelvic tumors (one case each of liver cancer, cervical cancer, and prostate cancer) admitted to Shandong Cancer Hospital and Institute from January to June 2024 were selected as the research subjects. IMPT and IMRT plans were designed for each case based on clinical target volume (CTV) and organs at risk (OARs) constraints. Dosimetric parameters, including conformity index (CI), homogeneity index (HI), and gradient index (GI) for target coverage, as well as OARs dose metrics, were evaluated. The volume of additional dose deposition in the body was compared by assessing regions receiving 10%, 30%, and 50% of the prescription dose.Results:For all three cases, IMRT plan demonstrated higher CI values (0.82, 0.81, and 0.86) compared to IMPT plan (0.61, 0.62, and 0.43). IMPT plan yielded lower HI values (0.053, 0.075, and 0.020) than IMRT plan (0.060, 0.120, and 0.080) and lower GI values (3.45, 2.63, and 3.80 vs. 7.28, 4.76, and 4.66 for IMRT plan). In liver cancer, IMPT plan reduced the D mean of normal liver tissues and right kidney by 37.8% and 78.5%, respectively, and decreased the D max of spinal cord by 13.2%. For cervical cancer, IMPT plan reduced the V 30 of the small bowel by 22.0%, D mean of the bladder, rectum and bone marrow by 15.7%, 14.3% and 12.6%, and spinal cord D max by 4.8%. In prostate cancer, IMPT plan lowered bladder and rectal D mean by 14.9% and 36.5%, respectively, but resulted in an increase of 35.3% and 6.1% in the D mean and V 40 of the left femoral head, respectively, and an increase of 23.6% and 10.8% in the D mean and V 40 of the right femoral head, respectively. IMPT plan reduced the volumes receiving 10%, 30%, and 50% of the prescription dose by 48.9%-64.8%, 22.0%-47.0%, and 22.0%-57.7%, respectively, compared to IMRT plan. Conclusions:Comparison between IMPT and IMRT plans for abdominopelvic tumors: IMPT plan offers advantages in reducing doses to normal organs such as the liver, kidneys, spinal cord, small intestine, rectum, and bladder. However, its advantage is less pronounced regarding the dose to the femoral heads. IMPT plan notably minimizes additional dose deposition within the body.

Objective:To investigate the dosimetric characteristics of intensity modulated proton therapy (IMPT) and photon volumetric modulated arc therapy (VMAT) in typical head and neck malignant tumors.Methods:Three types of typical head and neck tumors (nasopharyngeal carcinoma, parotid gland carcinoma, laryngeal carcinoma) treated at Shandong Cancer Hospital and Institute from December 2023 to December 2024 were taken as research subjects. IMPT and VMAT radiotherapy plans were created according to clinical prescription requirements of target and organs at risk limits respectively. The conformity index (CI), homogeneity index (HI) and gradient index (GI) for target coverage of two radiotherapy plans were evaluated for 3 patients, as well as the dosimetric indicators of organs at risk.Results:The CI of IMPT for nasopharyngeal carcinoma, parotid gland carcinoma and laryngeal carcinoma were 0.70, 0.72 and 0.67, respectively. The HI were 0.11, 0.08 and 0.08, respectively. The GI were 3.08, 2.49 and 3.75, respectively. The CI of VMAT plans were 0.77, 0.82 and 0.91, respectively. The HI were 0.12, 0.10 and 0.04, respectively. The GI were 3.67, 2.63 and 3.45, respectively. The results showed that CI of IMPT plan was slightly lower than that of VMAT plan, and HI of IMPT plan was comparable to that of VMAT plan, the GI of the IMPT plan for patients with nasopharyngeal carcinoma and parotid gland carcinoma was lower than that of the VMAT plan, and the GI of the IMPT plan for patient with laryngeal carcinoma was higher than that of the VMAT plan, and all were within the clinically acceptable range. The IMPT plan has demonstrated significant dose advantages in the treatment of nasopharyngeal carcinoma, parotid gland carcinoma and laryngeal carcinoma. For patient with nasopharyngeal carcinoma, the IMPT plan reduced the D max of the left and right crystals by 54.1% and 50.4%, respectively, compared to VMAT plan, and reduced the D mean of the oral and laryngeal tissues by 40.5% and 49.6%, respectively. For patient with parotid gland carcinoma, IMPT plan reduced the D max of the brainstem and spinal cord by 66.2% and 40.5%, respectively, compared to VMAT plan. For patient with laryngeal carcinoma, IMPT reduced spinal cord D max by 77.0%, while thyroid cartilage D mean increased by 8.0% compared to VMAT plan. For the additional dose in the patients' body, taking the absolute volumes occupied by the prescribed dose areas of 10%, 30%, and 50% in the patients' body as examples, IMPT plan of nasopharyngeal carcinoma patient decreased by 29.7%, 29.6%, and 34.9% compared to VMAT plan, respectively. IMPT plan of parotid gland carcinoma patient decreased by 61.0%, 39.7%, and 17.4% compared to VMAT plan, respectively. IMPT plan of laryngeal carcinoma patient decreased by 63.9%, 31.7%, and 4.1% compared to VMAT plan, respectively. Conclusions:Compared with VMAT plan, IMPT plan can effectively reduce the irradiation dose of most organs at risk near the target of head and neck tumors, but the dose of string organs close to the target area may be higher, which needs attention.

Objective:To investigate the dosimetric characteristics of intensity modulated proton therapy (IMPT) and intensity modulated radiation therapy (IMRT) for lung cancers.Methods:Three lung cancer patients (central-lower, central, and peripheral types) admitted to Shandong Cancer Hospital and Institute from January 2024 to May 2024 were selected as the research subjects. IMPT and IMRT plans were designed for each case based on the anatomical location of the clinical target volume and the dose constraints for organs at risk (OARs). Dosimetric parameters, including conformity index (CI), homogeneity index (HI), and gradient index (GI) for target coverage, as well as OARs dosimetric parameters were evaluated. The volume of additional dose deposition in the body was compared by assessing regions receiving 10%, 30%, and 50% of the prescription dose.Results:For all three cases, IMRT plans demonstrated higher CI values (0.80, 0.60, and 0.79) compared to IMPT plans (0.61, 0.57, and 0.34). IMPT plans yielded lower HI values (0.07, 0.06, and 0.06) than IMRT plans (0.09, 0.15, and 0.09) and lower GI values (2.84, 2.47, and 4.56 vs. 4.91, 3.09, and 4.99 for IMRT plans). Compared with the IMRT plans, the low-dose region in the ipsilateral lung was significantly reduced in IMPT plans (V 5 of the IMPT plans were 20.59%, 46.29%, 10.94%, respectively; V 5 of the IMRT plans were 48.91%, 60.63%, 19.92%, respectively), but there was no significant advantage in the high-dose region compared to IMRT plans (V 20 of the IMPT plans were 12.88%, 34.75%, 5.21%, respectively; V 20 of the IMRT plans were 21.70%, 36.50%, 5.31%, respectively). The dose to the contralateral lung and heart was significantly reduced in IMPT plans [the D mean of the contralateral lung in the IMPT plans were 0.08, 0.04, and 0.00 Gy (RBE), respectively, and those in the IMRT plans were 3.25, 1.18, and 0.55 Gy, respectively; the heart D mean in the IMPT plans were 6.23, 7.04, and 0.00 Gy (RBE), respectively, while those of the IMRT plans were 18.33, 10.27, and 0.08 Gy, respectively). IMPT plans significantly reduced the volumes receiving 10% of the prescription dose by 65.94%, 25.57% and 72.47%, respectively, compared to IMRT plans. The volumes IMPT plans occupied by 30% of the prescription dose area in the body were reduced by 54.97%, 26.47% and 39.04%, respectively, compared to the IMRT plans. The volumes IMPT plans occupied by 50% of the prescription dose area in the body were reduced by 54.49%, 30.43% and 28.89%, respectively, compared to the IMRT plans. Conclusions:IMPT plan significantly reduces the V 5 of the ipsilateral lung, the D mean of the contralateral lung and the heart, while maintaining target coverage compared with IMRT plan for lung cancers. However, IMPT plan does not show much more advantage than IMRT plan in the ipsilateral lung V 20. IMPT can reduce the additional exposure volume within the body.

Objective:To explore the dosimetric characteristics of proton radiotherapy plan and photon radiotherapy plan for esophageal cancer.Methods:Four patients who were admitted to Shandong Cancer Hospital and Institute from January 2024 to April 2024 with esophageal cancer (cervical, middle thoracic and total esophageal tube, as well as the lymphatic drainage areas involved) and required radiotherapy were selected as the research subjects. Intensity modulated proton therapy (IMPT) and intensity modulated radiation therapy (IMRT) plans were designed respectively based on the clinical target volume and the dose constraints for organs at risk (OARs). Dosimetric parameters, including conformity index (CI), homogeneity index (HI), gradient index (GI) for target coverage, as well as OARs dosimetric parameters were evaluated. The volume of additional dose deposition in the body was compared by assessing regions receiving 10%, 30%, and 50% of the prescription dose.Results:For all four cases, IMPT plans yielded lower HI values (0.12, 0.10, 0.06, and 0.08) than IMRT plans (0.15, 0.13, 0.10, and 0.11), and the GI values of IMPT plans (3.11, 3.21, 2.43, and 2.72) was lower than IMRT plans (4.52, 5.14, 3.09, and 3.92). Moreover, the CI of the IMPT plans (0.59, 0.60, 0.77, and 0.72) was inferior to IMRT plans (0.81, 0.77, 0.91, and 0.85). Compared with the IMRT plans, in the whole lung dose indicators of the IMPT plans for the 4 patients, V 5 decreased by 34.1%, 55.0%, 79.7% and 60.3%, respectively; V 20 decreased by 48.3%, 43.9%, 65.8% and 40.8%, respectively, and D mean decreased by 43.4%, 57.2%, 76.2% and 45.4%, respectively. V 30 of the heart decreased by 36.2%, 45.3%, 40.1% and 52.4%, respectively, and D mean of heart decreased by 96.6%, 57.9%, 58.5% and 55.3%, respectively. For the middle and lower thoracic target area, the liver was significantly protected in the IMPT plan (D mean decreased by 76.0% compared with the IMRT plan). In terms of the additional dose deposition in the patient's body, IMPT plans reduced the volumes receiving 10%, 30% and 50% of the prescription dose by 45.0%-61.4%, 41.2%- 61.8% and 34.8%-61.6%, respectively, compared with the IMRT plans. Conclusions:By comparing the dosimetric parameters of IMPT and IMRT plans for 4 cases of esophageal cancer, the IMPT plans have advantages in reducing the doses to lung tissue, heart, and liver, and can also reduce additional dose deposition in the patient's body.

Objective:To investigate the effect of mitochondrial division of GABAergic neurons in substantia nigra pars reticulata(SNr)on motor dysfunction in mice with acute hepatic encephalopathy(AHE).Methods:AHE mice model was established by intraperitoneal injection of thioacetamide(TAA).The changes of liver lobules in AHE mice were observed by hematoxylin-eosin(HE)staining.The changes of serum aspartate aminotransferase(AST),alanine aminotransferase(ALT)and blood ammonia in AHE mice were detected by biochemical detection kit.Then,the motor function of AHE mice was observed by rod fatigue test,elevated cross maze test and open field test.Furthermore,the changes of mitochondrial area,perimeter,roundness and other morphological indicators in SNr of AHE mice were ob-served and analyzed by transmission electron microscopy.The expression of mitochondrial division and fusion related molecules in SNr of AHE mice was observed by Western Blot.Then,the expression of mitochondrial dynamic related protein 1(DRP1)in SNr of AHE mice was targeted by AAV virus.The mitochondrial membrane potential(MMP),ATP and reactive oxygen species(ROS)in SNr were detected by fluorescence enzyme marker,and the changes of motor function of mice were observed.Results:Compared with the control group,the motor function of AHE mice was signifi-cantly decreased,the mitochondrial division of SNr was significantly enhanced,and the expression of mitochondrial divi-sion related proteins was significantly increased.The MMP in SNr of AHE mice was significantly decreased,the ATP of cells was decreased,and the ROS was increased.After targeted inhibition of DRP1 expression in SNr of AHE mice,the movement was improved;further observation found that after the mitochondrial division in SNr of AHE mice was inhibi-ted,the MMP was significantly increased,the ATP of cells was increased,and the ROS was decreased,which demon-strated that the mitochondrial function was significantly improved.Conclusion:Targeted inhibition of mitochondrial di-vision of GABAergic neurons in SNr of AHE mice can improve mitochondrial morphology and function,thus alleviating their movement disorders.

Abstract Socioeconomic status is an important factor affecting all-cause mortality. Income, education and occupation alone or in combination have been employed as a measure of socioeconomic status; however, the study results vary in measures. Material mechanism, lifestyle mechanism, psychological mechanism and community neighborhood mechanism have been accepted as the main intermediate mechanisms for the impact of socioeconomic status on all-cause mortality; however, the contribution of these mechanisms remains controversial. Based on the international and national publications pertaining to the association between socioeconomic status and all-cause mortality from 2012 to 2021, this review summarizes the relationship between socioeconomic status and all-cause mortality in different metrics and the intermediate mechanism of the impact of socioeconomic status on all-cause mortality, so as to provide insights for further studies.

Objective To explore the effect of serpine1, α-serpine1 protein and KLF2 in patients with lower extremity venous thrombosis by Fufang Xueshuantong capsule.Methods 44 patients were diagnosed with lower extremity venous thrombosis were collected, and randomly divided into experimental group and control group, 22 cases in each group, giving the conventional antiplatelet therapy, on the basis of this, the experimental group was given Fufang Xueshuantong capsule 1.5 g, three times a day orally, and the control group was given context Shutong granules 20 g, three times a day orally.2 weeks for a course of treatment, three courses of treatment.Before and after treatment in the detection of two groups of patients with serum serpine1, α-serpine1 KLF2 and protein contents, blood rheology, coagulation function, and carries on the analysis.ResuIts Compared with the control group, experimental group patients has better curative effect, the performance of: serum serpine1and α-serpine1 protein content were significantly decreased(P<0.05).KLF2 expression was significantly higher (P<0.05).Blood viscosity, erythrocyte aggregation index, erythrocyte sedimentation rate decreased significantly (P<0.05).Fib, D-D content significantly decreased (P<0.05).ConcIusions Fufang Xueshuantong capsule can reduce serum serpine1, α-serpine1 protein content and promote KLF2 expression, decrease the blood viscosity, erythrocyte sedimentation rate, improve patients with hypercoagulable state in patients with lower extremity venous thrombosis, so as to achieve good clinical efficacy.