Objective To evaluate the performance of lightweight Chinese large language models (LLMs) in answering specialized lung cancer questions, and to explore the impact of retrieval-augmented generation (RAG) on model performance. Methods Eleven lightweight Chinese LLMs with parameter sizes ranging from 7B to 32B were included. A lung cancer-specific evaluation dataset consisting of 200 questions [100 A1-type (basic knowledge) and 100 A2-type (clinical case) questions], constructed based on clinical guidelines and thoracic surgery textbooks, was used for assessment. Model performance was evaluated under two conditions (with and without RAG). Accuracy was used to assess model performance, and response latency was recorded to reflect inference efficiency. An accuracy–latency scatter plot was constructed for descriptive analysis of overall model performance. Results All models successfully completed the evaluation. With the introduction of RAG, the overall average accuracy improved from 61.68% to 76.36%. Smaller models demonstrated the most significant improvement (e.g., the accuracy of DeepSeek-7B increased from 32.50% to 60.00%, P<0.001). The average response latency increased from 12.58 s to 13.80 s. The Qwen3 series showed the best overall performance, and Qwen3-32B achieved the highest accuracy under both conditions (76.50% and 84.00%, respectively). After RAG integration, performance differences among model families were markedly reduced. Based on the accuracy-latency trade-off, Qwen3-32B achieved the best balance between accuracy and response latency under the baseline condition, whereas Qwen3-14B demonstrated superior overall performance in terms of accuracy, latency, and computational cost after RAG integration. Conclusion The integration of RAG technology improves the ability of lightweight Chinese LLMs to answer specialized lung cancer questions. Under the dual practical constraints of limited computational resources and medical data security requirements, the "lightweight model+RAG" technical framework may represent a promising deployment solution.

With the continuous development of digital technology, three-dimensional (3D) facial scanning technology is becoming more and more widely used in dentistry, where it enhances the precision of various processes of clinical diagnosis and treatment, and provides new perspectives and tools for dental research and application. Nowadays, 3D facial scanning technology meets clinical needs in terms of accuracy, and most facial scanners have an accuracy of 140-1 330 μm, and the 3D facial scanning related applications of smartphones can also meet the clinical application needs to a certain extent. In the clinical field, the 3D scanner can be combined with a variety of technologies to achieve multi-modal data fusion and create virtual patients, and its prediction rate in obstructive sleep apnea hypopnea syndrome is as high as 91%, which has great clinical application value. This article aims to explore the research status, clinical applications, and accuracy analysis of 3D facial scanning techniques in the field of dentistry.

With the continuous development of digital technology, three-dimensional (3D) facial scanning technology is becoming more and more widely used in dentistry, where it enhances the precision of various processes of clinical diagnosis and treatment, and provides new perspectives and tools for dental research and application. Nowadays, 3D facial scanning technology meets clinical needs in terms of accuracy, and most facial scanners have an accuracy of 140-1 330 μm, and the 3D facial scanning related applications of smartphones can also meet the clinical application needs to a certain extent. In the clinical field, the 3D scanner can be combined with a variety of technologies to achieve multi-modal data fusion and create virtual patients, and its prediction rate in obstructive sleep apnea hypopnea syndrome is as high as 91%, which has great clinical application value. This article aims to explore the research status, clinical applications, and accuracy analysis of 3D facial scanning techniques in the field of dentistry.

Objective:To evaluate the efficacy and safety of radiotherapy as a combined mode with in-situ vaccine for patients with advanced soft tissue sarcoma(STS).Methods:The clinical data of 12 patients with advanced STS who received combination therapy mode at the Cancer Center of Gulou Hospital Affiliated to the School of Medicine of Nanjing University between December 2020 and September 2024 were retrospectively analyzed.All 12 patients received combined therapy.The main radiotherapeutic approach was hypofractionated radiotherapy.The targeted therapy mainly involved Anlotinib(in 10 cases)or Apatinib(in 2 cases).Immunotherapy mainly involved PD-1 antibodies.The primary endpoint was disease control rate(DCR),and the secondary endpoints were objective response rate(ORR)and safety.Results:Among the 12 STS patients who received combined treatment,0 cases achieved CR,4 cases achieved PR;7 cases had SD,and 1 case had PD.The ORR was 33%,and the DCR was 91.7%,among which the DCR of the target lesions was 100%.Among the 12 patients,9 patients experienced grade Ⅰ to grade Ⅱ adverse reactions.The most frequently occurring hematological adverse reactions were anemia(6 cases)and abnormal results of liver function tests(3 cases).The most frequently occurring non-hematological adverse reactions were proteinuria(5 cases),hypertension(4 cases),abnormal thyroid function(3 cases),anorexia(3 cases),and nausea and vomiting(2 cases).Only 2 cases had grade Ⅲ hematological toxicity,and 1 case had grade Ⅲpneumothorax.Conclusion:Radiotherapy as a combined therapy mode with in situ vaccine can achieve a higher DCR in advanced soft tissue sarcomas without serious adverse reactions.This combined treatment modality demonstrates good efficacy and safety.

Objective:To assess the feasibility of using the posterosuperior tetralogy (PS-Tetra) score for predicting the prognosis of repair of posterosuperior massive rotator cuff tears.Methods:Data were retrospectively reviewed for patients who underwent repair of posterosuperior massive rotator cuff tears from February 2016 to June 2020. A total of 95 (male 48, female 47) shoulders with an average age of 58.52±8.33 years (range, 27-76 years) were included. The American Shoulder and Elbow Surgeons (ASES) scores and shoulder range of motions (ROM) were used to evaluate shoulder function. MRI was used to assess preoperative fatty infiltration (FI), atrophy, modified Patte's classification, PS-Tetra score and postoperative tendon integrity. The shoulder function was compared between groups of different PS-Tetra scores. The binary logistic regression was used to determine the risk factors of irreparability and retear.Results:83 cases of repair of posterosuperior massive rotator cuff tears were finally included. Complete repairs were performed in 83 cases, and partial repairs were performed in 12 cases. Retear was observed in 17 (20%) cases. The ASES scores (postoperative 58.52±8.33 vs. preoperative 47.30±17.40, t=-19.642, P<0.001), ROM of forward flexion (postoperative 157.60°±13.85° vs. preoperative 116.88°±50.89°, t=-7.272, P<0.001), external rotation (postoperative 45.26°±14.69° vs. preoperative 37.34°±18.65°, t=-4.043, P<0.001) and internal rotation [postoperative L 1 (T 7-buttock) vs. preoperative L 2 (T 7-buttock), Z=-2.737, P=0.006] were significantly improved postoperatively in the group with PS-Tetra score between 0 and 2. In the group with PS-Tetra score of 3 and 4, the ASES scores (postoperative 69.17±15.91 vs. preoperative 46.85±20.73, t=-11.167, P=0.001) were significantly improved postoperatively, while the ROMs were not. Modified Patte stageⅢ[ OR=26.827, 95% CI (2.089, 344.500), P=0.012] was the risk factor of irreparability. Dominant side involvement [ OR=9.407, 95% CI (1.044, 84.784), P=0.046) and PS-Tetra score of 3 and 4 [ OR=5.037, 95% CI (1.028, 26.623), P=0.046] were risk factors of retear. Conclusions:For repair of posterosuperior massive rotator cuff tears, preoperative PS-Tetra score of 3 and 4 was the risk factors of poor postoperative shoulder functions and retear.

Objective To investigate the dietary patterns of rural residents in the high-incidence areas of esophageal cancer (EC), and to explore the clustering and influencing factors of risk factors associated with high-incidence characteristics. Methods A special structured questionnaire was applied to conduct a face-to-face survey on the dietary patterns of rural residents in Yanting county of Sichuan Province from July to August 2021. Univariate and multivariate logistic regression models were used to analyze the influencing factors of risk factor clustering for EC. Results There were 838 valid questionnaires in this study. A total of 90.8% of rural residents used clean water such as tap water. In the past one year, the people who ate fruits and vegetables, soybean products, onions and garlic in high frequency accounted for 69.5%, 32.8% and 74.5%, respectively; the people who ate kimchi, pickled vegetables, sauerkraut, barbecue, hot food and mildew food in low frequency accounted for 59.2%, 79.6%, 68.2%, 90.3%, 80.9% and 90.3%, respectively. The clustering of risk factors for EC was found in 73.3% of residents, and the aggregation of two risk factors was the most common mode (28.2%), among which tumor history and preserved food was the main clustering pattern (4.6%). The logistic regression model revealed that the gender, age, marital status and occupation were independent influencing factors for the risk factors clustering of EC (P<0.05). Conclusion A majority of rural residents in high-incidence areas of EC in Yanting county have good eating habits, but the clustering of some risk factors is still at a high level. Gender, age, marital status, and occupation are influencing factors of the risk factors clustering of EC.

Objective:To explore the construction and application of the standard process of thirst management for patients undergoing endonasal transsphenoidal surgery for pituitary adenomas.Methods:From June 2020 to October 2021, 160 patients with endonasal transsphenoidal surgery for pituitary adenomas who were admitted to the Neurosurgery Department of the First Affiliated Hospital of University of Science and Technology of China (Anhui Provincial Hospital) were selected by convenience sampling. The patients were randomly divided into test group and control group, 80 cases in each group. The control group implemented routine nursing to prevent postoperative thirst, while the test group established a standard process of thirst management and implemented thirst prevention and interventions according to the process. The thirst intensity, thirst pain and resting salivary flow rate between the two groups was compared.Results:The thirst intensity scores of patients in the test group at 2, 6 and 24 hours after operation were lower than those in the control group, with statistically significant differences ( Z=-8.696, -10.008, -4.305; P<0.01) . The score of thirst pain in the test group was lower than that in the control group, and the resting salivary flow rate at 6 and 24 hours after operation were higher than those in the control group, with statistically significant differences ( Z=-9.488, -4.124, -2.631; P<0.01) . Conclusions:The standard process of thirst management can reduce the postoperative thirst intensity of patients with pituitary adenomas surgery and improve their thirst feeling, which is worthy of clinical promotion.

OBJECTIVE: To evaluate the effectiveness of tendon insertion medialized repair in treatment of large-to-massive rotator cuff tears (L/MRCT). METHODS: The clinical and imaging data of 46 L/MRCT patients who underwent arthroscopic insertion medialized repair between October 2015 and June 2019 were retrospectively analyzed. There were 26 males and 20 females with an average age of 57.7 years (range, 40-75 years). There were 20 cases of large rotator cuff tears and 26 cases of massive rotator cuff tears. Preoperative imaging evaluation included fatty infiltration (Goutallier grade), tendon retraction (modified Patte grade), supraspinatus tangent sign, acromiohumeral distance (AHD), and postoperative medializaiton length and tendon integrity. The clinical outcome was evaluated by visual analogue scale (VAS) score, American Society for Shoulder and Elbow Surgery (ASES) score, shoulder range of motion (including anteflexion and elevation, lateral external, and internal rotation) and anteflexion and elevation muscle strength before and after operation. The patients were divided into two groups (the intact tendon group and the re-teared group) according to the integrity of the tendon after operation. According to the medializaiton length, the patients were divided into group A (medialization length ≤10 mm) and group B (medialization length >10 mm). The clinical function and imaging indexes of the patients were compared. RESULTS: All patients were followed up 24-56 months, with an average of 31.8 months. At 1 year after operation, MRI showed that the medializaiton length of supraspinatus tendon was 5-15 mm, with an average of 10.26 mm, 33 cases in group A and 13 cases in group B. Eleven cases (23.91%) had re-teared, including 5 cases (45.45%) of Sugaya type Ⅳ and 6 cases (54.55%) of Sugaya type Ⅴ. At last follow-up, the VAS score, ASES score, shoulder anteflexion and elevation range of motion, lateral external rotation range of motion, and anteflexion and elevation muscle strength significantly improved when compared with those before operation ( P<0.05); there was no significant difference in internal rotation range of motion between pre- and post-operation ( P>0.05). The Goutallier grade and modified Patte grade of supraspinatus muscle in the re-teared group were significantly higher than those in the intact tendon group, and the AHD was significantly lower than that in the intact tendon group ( P<0.05). There was no significant difference in other baseline data between the two groups ( P>0.05). Except that the ASES score of the intact tendon group was significantly higher than that of the re-teared group ( P<0.05), there was no significant difference in the other postoperative clinical functional indicators between the two groups ( P>0.05). There was no significant difference in the incidence of re-tear, VAS score, ASES score, range of motion of shoulder joint, and anteflexion and elevation muscle strength between group A and group B ( P>0.05). CONCLUSION Tendon insertion medialized repair may be useful in cases with L/MRCT, and shows good postoperative shoulder function. Neither tendon integrity nor medialization length shows apparent correlations with postoperative shoulder function.
Male ; Female ; Humans ; Middle Aged ; Rotator Cuff Injuries/surgery* ; Retrospective Studies ; Treatment Outcome ; Rotator Cuff/surgery* ; Tendons ; Rupture/surgery* ; Shoulder Joint/surgery* ; Arthroscopy/methods* ; Range of Motion, Articular

Objective:To develop the Peripherally Inserted Central Catheter (PICC) Nursing Knowledge, Attitude and Practice Scale for Primary Caregivers of Patients with PICC, and to test its reliability and validity.Methods:Taking the theory of knowledge, attitude and practice as the basic framework, the item pool was initially constructed through literature review, group discussion, and semi-structured interviews from March to April 2021. From May to July 2021, the initial scale was formed through Delphi expert consultation, and the items of the scale were pre-tested and adjusted through cognitive interviews. From June to October 2021, convenience sampling was used to select the primary caregivers of 450 patients with PICC in a cancer hospital in Hunan Province for a formal survey to test the reliability and validity of the scale.Results:The formal scale consisted of 53 items in 3 dimensions, including 19 items in the knowledge dimension, 8 items in the attitude dimension, and 26 items in the practice dimension. The Cronbach's α coefficients of each dimension and the total scale were 0.949, 0.962, 0.972, and 0.981, respectively, and the test-retest reliability coefficients of each dimension and the total scale were 0.951, 0.935, 0.972, and 0.967, respectively. The split-half reliability coefficients of each dimension and the total scale were 0.889, 0.954, 0.889, and 0.882, respectively. The content validity index of the item ranged from 0.82 to 1.00.Conclusions:The reliability and validity of the PICC Nursing Knowledge, Attitude and Practice Scale for Primary Caregivers of Patients with PICC is good, and it can be used as a tool to evaluate the PICC nursing knowledge, attitude, and practice of primary caregivers of patients with PICC.

Objective:To investigate the role and regulatory mechanism of triggering receptor expressed on myeloid cell 2 (TREM2) in mice lung ischemia/reperfusion injury (LIRI).Methods:Thirty-six healthy male C57BL/6 mice were divided into six groups according to the random number method ( n = 6): normal control group, and LIRI 2, 6, 12, 24, 48 hours group. Mice LIRI models were established by clamping the left hilum. The wet/dry weight ratio (W/D) of left lung tissue was measured. Lung injury was observed and evaluated by hematoxylin-eosin (HE) staining and electron microscopy. The levels of interleukins (IL-1β, IL-18) in lung tissue were detected by enzyme linked immunosorbent assay (ELISA). The mRNA expressions of TREM2 and caspase-1 were determined by polymerase chain reaction (PCR). The protein expressions of TREM2, caspase-1, Gasdermin-D (GSDMD) were determined by Western blotting. Results:At 2 hours after LIRI, lung injury began to appear, the lung ultrastructure changed, and the lung injury score increased; at 6 hours, the degree of lung injury was the most serious; after 12 hours, the lung injury gradually reduced and the lung injury score gradually decreased. Compared with the normal control group, lung W/D ratio and lung injury score of LIRI 2, 6, 12, 24, 48 hours groups were significantly higher, the differences were statistically significant (lung W/D ratio: 7.06±0.52, 8.34±0.17, 6.42±0.35, 5.34±0.25, 5.59±0.45 vs. 4.69±0.23; lung injury score: 5.50±0.54, 9.75±0.89, 5.88±0.84, 3.63±0.74, 4.13±0.64 vs. 1.13±0.35, all P < 0.05). Compared with the normal control group, the levels of IL-1β and IL-18 in lung tissue were significantly increased at 2 hours after LIRI, reached a peak at 6 hours [IL-1β (ng/L): 502.76±12.25 vs. 56.50±8.07, IL-18 (ng/L): 414.02±10.75 vs. 81.63±5.29, both P < 0.05], then decreased gradually, and were still significantly higher than the normal control group at 48 hours. The PCR and Western blotting showed that the expression of TREM2 was significantly lower than that in the normal control group at 2 hours after LIRI, and reached a valley at 6 hours [TREM2 mRNA (2 -ΔΔCt): 0.47±0.05 vs. 1.02±0.05, TREM2/GAPDH: 0.23±0.13 vs. 0.48±0.17, both P < 0.05], then gradually increased, and reached the peak at 24 hours [TREM2 mRNA (2 -ΔΔCt): 3.98±0.15 vs. 1.02±0.05, TREM2/GAPDH: 0.71±0.17 vs. 0.48±0.17, both P < 0.05]. The trend of expression of caspase-1 and GSDMD were opposite to that of TREM2, which increased at first and then decreased, and reached a peak at 6 hours after reperfusion [caspase-1 mRNA (2 -ΔΔCt): 2.20±0.13 vs. 1.01±0.02, caspase-1/GAPDH: 0.64±0.02 vs. 0.20±0.06, GSDMD/GAPDH: 1.23±0.01 vs. 0.87±0.02, all P < 0.05]. Conclusions:TREM2 might be involved in LIRI in mice. The mechanism may be related to the effect of TREM2 on caspase-1-mediated pyroptosis.