With the continuous deepening of the concept of precision diagnosis and treatment for lung cancer, how to achieve higher efficiency and accuracy in the screening, diagnosis, and treatment pathways in clinical practice has become an important issue that urgently needs to be overcome. The current clinical difficulty lies in the fact that despite continuous advancements in imaging and molecular diagnostic technologies, there are still limitations in manual efficiency and subjective experience when it comes to massive data analysis and multi-scale feature extraction. Artificial intelligence (AI), especially algorithm systems based on deep learning, is an innovative technology capable of deeply empowering medical big data. This method utilizes algorithms such as convolutional neural networks, combined with radiomics, pathomics, and multi-modal data fusion analysis, demonstrating immense potential in early precise detection and benign-malignant differentiation of pulmonary nodules, digital pathological subtype recognition and non-invasive prediction of driver genes, precise 3D surgical planning and automatic delineation of radiotherapy target volumes, as well as dynamic risk warning during follow-up. This innovative technology provides a brand-new solution for realizing intelligent and individualized lung cancer diagnosis and treatment models. This consensus, based on the latest evidence from evidence-based medicine and combined with the development trends in the AI field and real-world clinical needs, was ultimately formed by gathering the consensus opinions of multidisciplinary experts in radiology, pathology, thoracic surgery, and other fields. The main content covers the application specifications of AI in the three core scenarios of lung cancer screening, diagnosis, and treatment, the technical standards for data collection and algorithm validation, as well as the ethical and regulatory challenges faced at the current stage. It aims to clarify the applicable boundaries of AI as a clinical auxiliary decision support tool, providing scientific guidance and standardized exploration directions for peers currently engaged in or planning to carry out AI-assisted clinical diagnosis, treatment, and translation of lung cancer.

ObjectiveTo establish an objective evaluation system based on the quality characteristics of Gardeniae Fructus without relying on the origin label according to the theory of excellent shape， high quality， and superior effect of authentic medicinal materials， thus providing a scientific basis for the standardization of Gardeniae Fructus. MethodsDifferent batches of Gardeniae Fructus samples were collected from different producing areas. Through herbal textual research and a review of modern literature， fruit length， maximum fruit diameter， and chroma were selected as quantitative indicators to characterize the traditional appearance traits of Gardeniae Fructus. The above indicators were measured by a vernier caliper and a spectrophotometer. Through classical famous prescription data and literature analysis， total iridoid glycosides， total crocus glycosides， and total flavonoids were selected as the core component indicators related to efficacy， and their content was determined. Finally， through cluster analysis， the threshold ranges of the quantitative indicators of excellent shape and high quality were determined， and the applicability of the established evaluation system was verified by the determination of the indicators for each batch of medicinal materials. ResultsThe excellent shape indicators of Gardeniae Fructus were the fruit length-to-maximum fruit diameter ratio ≤1.86， 0°≤h≤45°， and L^*≥ 37.32. The content of total iridoid glycosides， total crocus glycosides， and total flavonoids should be no less than 2.2%， 0.44%， and 0.028%， respectively. The measurement results of different batches of Gardeniae Fructus showed that the evaluation system could effectively distinguish the Gardeniae Fructus samples in line with the traditional cognition of excellent shape and high quality. ConclusionThis study preliminarily constructs an excellent shape and high quality-based quality evaluation system for Gardeniae Fructus， which links appearance traits with chemical indicators and provides a methodological basis for the quality control and scientific grading of this herbal medicine.

Lung cancer constitutes the most prevalent and lethal malignant tumor in China. Approximately 85% of lung cancer diagnoses correspond to the non-small cell histological subtype [non-small cell lung cancer (NSCLC)]. Despite surgery being the mainstay for early-stage disease, postoperative recurrence remains high and adjuvant chemotherapy offers limited benefit. In recent years, targeted therapy has demonstrated substantial advantages in driver mutation-positive NSCLC. To this end, the Lung Cancer Medical Education Committee of the Chinese Medical Education Association developed guidelines based on a systematic review of evidence through November 2025, using the Grading of Recommendations, Assessment, Development and Evaluations (GRADE) approach and a modified Delphi method. Focusing on epidermal growth factor receptor (EGFR) and anaplastic lymphoma kinase (ALK), and addressing ROS proto-oncogene 1 (ROS1), B-Raf proto-oncogene serine/threonine kinase (BRAF) V600E mutation, and mesenchymal-epithelial transition factor (MET) exon 14 (METex14) skipping, the guideline covers molecular testing, neoadjuvant/adjuvant therapy, perioperative strategies, minimal residual disease monitoring, and postoperative surveillance. It defines testing requirements, specifies stage-directed and subtype-specific treatments, and standardizes minimal residual disease monitoring. These recommendations emphasize precision and feasibility to improve survival and quality of life.

ObjectiveTo develop and evaluate an O-RADSbased multiparametric model integratingcontrast-enhanced US （CEUS） and the z-score standardized risk of ovarian malignancy algorithm （ROMA z-score） to improve the diagnostic accuracy for differentiating benign， borderline， and malignant ovarian epithelial tumors （OETs）. MethodsThis retrospective study included 129 patients with pathologically confirmed OETs between May 2018 and December 2024. Clinical variables （age， menopausal status， BMI， HE4， CA125） and sonographic features （conventional US and CEUS） were collected. Tumors were classified as benign， borderline， or malignant according to pathological findings. A random forest algorithm was used to contruct the predictive model. Receiver operating characteristic （ROC） curves were generated tocompare the diagnostic performance of O-RADS， O-RADS+CEUS， O-RADS+ROMA z-score， and O-RADS+ CEUS+ROMA z-score. ResultsAmong the 129 patients， 63 had benign tumors， 25 had borderline tumors， and 41 had malignant tumors. Compared with O-RADS alone， the combined O-RADS+CEUS+ROMA z-score model demonstrated superior diagnostic performance， with an AUC of 0.87 （95% CI： 0.79-0.95） vs. 0.64 （95% CI： 0.51-0.76）， P＜0.001， for differentiating benign from borderline tumors， and 0.90 （95% CI： 0.83-0.97） vs. 0.64 （95% CI： 0.50-0.78）， P＜0.001， for differentiating borderline from malignant tumors. The random forest analysis identified HE4， CA125 and CEUS features as the most important predictors，contributing 11.02， 8.82， 18.1 points， respectively， to the total importance score of 51.94. In the test cohort， the model achieved an overall diagnostic accuracy of 72.97%， with an AUC of 0.89（95%CI：0.77-1.00）， sensitivity of 67.06%， and specificity of 85.01%. ConclusionsO-RADS+CEUS+ROMA z-score multiparametric model demonstrated the highest diagnostic performance for in differentiating benign from borderline and borderline from malignant OETs. HE4， CA125 and CEUS features were identified as key predictive risk factors， supporting the role of this model in improving preoperative risk stratification of OETs.

ObjectiveTo enhance the genetic diagnostic yield for fetal short femur length （FL） and to evaluate the integration of whole exome sequencing （WES） with chromosomal abnormality detection in prenatal diagnosis. MethodsA total of 27 mid-to-late trimester fetuses diagnosed with short FL ［FL < 5th percentile for gestational age or < -2 SD （Standard Deviation）］ at the Prenatal Diagnosis And Medical Genetics Center，Shenshan Medical Center，Memorial Hospital of Sun Yat-Sen University between November 2023 and May 2025 were included. Amniotic fluid or umbilical cord blood samples were collected and subjected to chromosomal karyotyping （19cases）， chromosomal microarray analysis （CMA） or low-coverage whole genome copy number variation（CNV-Seq） （27 cases）， and trio-based whole exome sequencing （trio-WES） （27 cases）. The pathogenicity of genetic variants were interpreted according to the American College of Medical Genetics and Genomics （ACMG） guidelines. ResultsTwenty-six genomic variants were detected in 17 of the 26 fetuses， including numerical chromosomal abnormalities （mosaic type）， copy number variations （CNVs）， single nucleotide variants （SNVs） and insertions-deletions （INDELs）. Genomic variants detected in 11 cases could explain the short femur phenotype.The overall diagnostic rate was 11/27， with pathogenic/likely pathogenic variants explaining 9 cases and variants of uncertain significance （VUS） accounting for 1 cases. In addition Chromosomal karyotype analysis revealed chromosome mosaicism explaining 1 case. The diagnostic rate of CNV detection （CMA/CNV-seq） was 2/27， while WES achieved a diagnostic rate of 10/27， with an incremental yield of 8/27 in cases with negative or inconclusive CNV results. Pathogenic mutations involved genes such as FGFR3， SHOX， DUOX2， CLCN5， and HBA1/HBA2. ConclusionTrio-WES significantly improves the genetic diagnostic rate for fetal short femur length， particularly in cases where chromosomal analysis fails to identify a cause. Karyotyping remains valuable for detecting low-level mosaicism in large-scale chromosomal abnormalities， thereby complementing WES and CNV detection. A stepwise or parallel genetic testing strategy is recommended in prenatal diagnosis to optimize diagnostic efficiency and support clinical decision-making.

OBJECTIVE To provide a reference for the optimization of prescription-review software and the selection of auxiliary prescription-review tools for medical institutions.METHODS A total of 559 eligible prescriptions were collected from the Meikang Prescription Automatic Screening System of a tertiary grade-A hospital from January 2025 to March 2026. Judged according to the composite reference standard （CRS） established in this study， there were 206 positive prescriptions [with drug-drug interaction （DDI）] and 353 negative prescriptions （without DDI）. DeepSeek， Doubao， Wenxin yiyan， Kimi and Tongyi qianwen were separately adopted to identify DDI in the 559 prescriptions， and binary judgment was performed on their output texts related to DDI. Taking the results judged by CRS as the gold standard， the confusion matrix was constructed and model performance indicators were calculated. The differences in sensitivity between any two large language models were compared， the relative sensitivity （RS） and its 95% confidence interval （CI） were calculated， and the results were visualized by a forest plot.RESULTS Wenxin yiyan achieved the highest sensitivity （74.8%） with the lowest risk of missed reports， whereas its specificity was only 40.8%. It had the lowest accuracy （53.3%） and presented a high risk of false positive reports. Tongyi qianwen obtained the highest accuracy （65.3%） and the optimal F1-score （0.587）. DeepSeek showed moderate performance in both sensitivity and specificity. Doubao and Kimi exhibited relatively low sensitivity （38.8% and 45.1%， respectively）， accompanied by high risks of missed reports. Pairwise comparison revealed that the sensitivity of Wenxin yiyan was significantly higher than that of DeepSeek [ χ 2 ＝4.84， P ＝0.028； RS＝1.124， 95%CI （1.018， 1.230）]. The sensitivity of Tongyi qianwen was significantly higher than that of Doubao [ χ 2 ＝4.77， P ＝0.029； RS＝1.239， 95%CI （1.035， 1.443）]. DeepSeek had significantly higher sensitivity than Doubao [ χ 2 ＝37.65， P ＜0.001； RS＝1.715， 95%CI （1.491， 1.939）]. No statistically significant differences in sensitivity were observed between Kimi versus Doubao and between Tongyi qianwen versus Kimi （ P ＞0.05）.CONCLUSIONS Wenxin yiyan is applicable for preliminary-screening scenarios with low tolerance for missed reports and requires combined manual review. Tongyi qianwen is suitable for auxiliary outpatient prescription review. DeepSeek can act as a transitional solution for medical institutions without dedicated prescription-review software. Doubao and Kimi are not recommended for independent application in prescription DDI audit.

Objective: To evaluate the feasibility of dynamic contrast-enhanced MRI (DCE-MRI) in differentiating tumor deposits (TDs) from metastatic lymph nodes (MLNs) in rectal cancer. Materials and Methods: A retrospective analysis was conducted on 70 patients with rectal cancer, including 168 lesions (70 TDs and 98 MLNs confirmed by histopathology), who underwent pretreatment MRI and subsequent surgery between March 2019 and December 2022. The morphological characteristics of TDs and MLNs, along with quantitative parameters derived from DCE-MRI (K trans , kep, and v e) and DWI (ADCmin, ADCmax, and ADCmean), were analyzed and compared between the two groups.Multivariable binary logistic regression and receiver operating characteristic (ROC) curve analyses were performed to assess the diagnostic performance of significant individual quantitative parameters and combined parameters in distinguishing TDs from MLNs. Results: All morphological features, including size, shape, border, and signal intensity, as well as all DCE-MRI parameters showed significant differences between TDs and MLNs (all P < 0.05). However, ADC values did not demonstrate significant differences (all P > 0.05). Among the single quantitative parameters, v e had the highest diagnostic accuracy, with an area under the ROC curve (AUC) of 0.772 for distinguishing TDs from MLNs. A multivariable logistic regression model incorporating short axis, border, v e, and ADC mean improved diagnostic performance, achieving an AUC of 0.833 (P = 0.027). Conclusion The combination of morphological features, DCE-MRI parameters, and ADC values can effectively aid in the preoperative differentiation of TDs from MLNs in rectal cancer.

Objective: To evaluate the feasibility of dynamic contrast-enhanced MRI (DCE-MRI) in differentiating tumor deposits (TDs) from metastatic lymph nodes (MLNs) in rectal cancer. Materials and Methods: A retrospective analysis was conducted on 70 patients with rectal cancer, including 168 lesions (70 TDs and 98 MLNs confirmed by histopathology), who underwent pretreatment MRI and subsequent surgery between March 2019 and December 2022. The morphological characteristics of TDs and MLNs, along with quantitative parameters derived from DCE-MRI (K trans , kep, and v e) and DWI (ADCmin, ADCmax, and ADCmean), were analyzed and compared between the two groups.Multivariable binary logistic regression and receiver operating characteristic (ROC) curve analyses were performed to assess the diagnostic performance of significant individual quantitative parameters and combined parameters in distinguishing TDs from MLNs. Results: All morphological features, including size, shape, border, and signal intensity, as well as all DCE-MRI parameters showed significant differences between TDs and MLNs (all P < 0.05). However, ADC values did not demonstrate significant differences (all P > 0.05). Among the single quantitative parameters, v e had the highest diagnostic accuracy, with an area under the ROC curve (AUC) of 0.772 for distinguishing TDs from MLNs. A multivariable logistic regression model incorporating short axis, border, v e, and ADC mean improved diagnostic performance, achieving an AUC of 0.833 (P = 0.027). Conclusion The combination of morphological features, DCE-MRI parameters, and ADC values can effectively aid in the preoperative differentiation of TDs from MLNs in rectal cancer.