Lung cancer is currently one of the most aggressive malignant tumors in the world,and ra-diomics as an emerging technology can transform the image features into the quantifiable and computable data features,plays a crucial role in the diagnosis and treatment process of lung cancer.With the construction of medical big data,artificial intelligence(AI)and radiomics can make use of imaging images to provide more valuable information in screening and diagnosis,histopathological assessment and prediction of treatment re-sponse in lung cancer,thus assisting in the formulation of individualized treatment regimen,realizing accurate diagnosis and treatment and improving the patient's prognosis.However,there are still many problems to be solved in the application of AI systems to clinical practice.This paper reviews the research progress of AI and radiomics in lung cancer screening and diagnosis,staging and grading,and efficacy assessment,and discusses the limitations and challenges that still exist in the clinical practice in order to provide reference for subsequent research.

Abdominal aortic aneurysm (AAA) is a deadly condition of the aorta, carrying a significant risk of death upon rupture. Currently, there is a dearth of efficacious pharmaceutical interventions to impede the advancement of AAA and avert it from rupturing. Here, we investigated dihydromyricetin (DHM), one of the predominant bioactive flavonoids in Ampelopsis grossedentata (A. grossedentata), as a potential agent for inhibiting AAA. DHM effectively blocked the formation of AAA in angiotensin II-infused apolipoprotein E-deficient (ApoE^-/-) mice. A combination of network pharmacology and whole transcriptome sequencing analysis revealed that DHM's anti-AAA action is linked to heme oxygenase (HO)-1 (Hmox-1 for the rodent gene) and hypoxia-inducible factor (HIF)-1α in vascular smooth muscle cells (VSMCs). Remarkably, DHM caused a robust rise (∼10-fold) of HO-1 protein expression in VSMCs, thereby suppressing VSMC inflammation and oxidative stress and preserving the VSMC contractile phenotype. Intriguingly, the therapeutic effect of DHM on AAA was largely abrogated by VSMC-specific Hmox1 knockdown in mice. Mechanistically, on one hand, DHM increased the transcription of Hmox-1 by triggering the nuclear translocation and activation of HIF-1α, but not nuclear factor erythroid 2-related factor 2 (NRF2). On the other hand, molecular docking, combined with cellular thermal shift assay (CETSA), isothermal titration calorimetry (ITC), drug affinity responsive target stability (DARTS), co-immunoprecipitation (Co-IP), and site mutant experiments revealed that DHM bonded to HO-1 at Lys243 and prevented its degradation, thereby resulting in considerable HO-1 buildup. In summary, our findings suggest that naturally derived DHM has the capacity to markedly enhance HO-1 expression in VSMCs, which may hold promise as a therapeutic strategy for AAA.

Objective To observe the correlations of image manifestations of organ damage and clinical characteristics of heat stroke.Methods Totally 46 patients with heat stroke were retrospectively enrolled.Based on the presence or absence of image manifestations of heat stroke related organ damage on head MRI and chest-abdominal CT and combined with sequential organ failure assessment(SOFA),the patients were divided into organ damage positive group(positive A,images showed clear manifeatations of organ damage and the corresponding SOFA≥2)and organ damage negative group(negative B,images showed no obvious organ damage or the corresponding SOFA＜2).Clinical data,laboratory indicators and treatment outcomes were compared between groups.Univariate and multivariate logistic analyses were used to explore the correlations of image manifestations of organ damage and clinical characteristics of heat stroke.Results Among 46 cases,26 were classified as positive group A and 20 as negative group.There were significant differences of patients' age,gender,Glasgow coma scale(GCS)and heart rate at admission,thrombin time,D-dimer,lactate dehydrogenase,partial pressure of carbon dioxide,lactic acid,myoglobin,platelet count(PLC),interleukin-6 and serum creatinine before treatment,as well as time of hospital stay and post-treatment outcomes between groups(all P＜0.05).GCS score,PLC levels and time of hospital stay of heat stroke patients were all correlated with image manifeations of organ damage(OR=0.592,0.729,1.532,all P＜0.05),among which GCS score and time of hospital stay were independently associated with image manifeations of organ damage(OR=0.607,1.448,both P＜0.05).Conclusion Image manifeations of organ damage was correlated with GCS score and time of hospital stay in patients with heat stroke.

Objective:To develop CNN-Dinov2, a deep learning-based automatic classification model for Gram-stained images, enabling rapid diagnosis of three prevalent Gram-negative bacilli in bloodstream infections: Escherichia coli ( E.coli), Klebsiella pneumoniae ( K.pneumoniae), and Pseudomonas aeruginosa ( P.aeruginosa). Methods:This evaluation study analyzed 1 425 Gram-stained microscopic images from patients with bloodstream infections at Houjie Hospital, in Dongguan City, collected between January 2023 and January 2024. The images, all positive for blood culture and identified as target strains, were categorized into Escherichia coli (419 images), Klebsiella pneumoniae (411 images), Pseudomonas aeruginosa (413 images), and other Gram-negative bacilli (182 images). They were randomly split into a training set (1 141 images), a validation set (141 images), and a test set (143 images) in an 8∶1∶1 ratio. A hybrid CNN-Dinov2 model was developed by integrating ResNet′s local feature extraction with Dinov2′s global pre-trained features, followed by a fully connected layer. The model was optimized by inputting the preprocessed images and adjusting parameters through loss calculation and backpropagation. AlexNet, Dinov2, and ResNet18 served as control models. The models′ classification performance was assessed using accuracy, precision, weighted F1 score, and recall rate, derived from the confusion matrix. The PR curve and AP value further evaluated each model′s classification capability across the four image categories. Results:The CNN-Dinov2 model achieved a training accuracy of 99.74%, a validation accuracy of 98.12%, and a validation loss of 0.070 6, demonstrating robust generalization without overfitting. Validation metrics revealed superior performance with an accuracy of 98.60%, precision of 98.65%, a weighted F1 score of 98.60%, and a recall rate of 98.60%, outperforming other models. The confusion matrix confirmed its strong classification capability, with the highest sum of diagonal values for identifying four types of bacteria. The macro average precision (AP) values under the precision-recall (PR) curves were all 1, indicating excellent discrimination across all categories. Overall, the CNN-Dinov2 model exhibited the best performance among the four models evaluated.Conclusion:This study successfully developed CNN-Dinov2, an automated classification model for Gram staining images. It offers valuable support for the rapid diagnosis of bloodstream infections caused by Escherichia coli, Klebsiella pneumoniae, and Pseudomonas aeruginosa, demonstrating practical utility.

Objective To observe the correlations of image manifestations of organ damage and clinical characteristics of heat stroke.Methods Totally 46 patients with heat stroke were retrospectively enrolled.Based on the presence or absence of image manifestations of heat stroke related organ damage on head MRI and chest-abdominal CT and combined with sequential organ failure assessment(SOFA),the patients were divided into organ damage positive group(positive A,images showed clear manifeatations of organ damage and the corresponding SOFA≥2)and organ damage negative group(negative B,images showed no obvious organ damage or the corresponding SOFA＜2).Clinical data,laboratory indicators and treatment outcomes were compared between groups.Univariate and multivariate logistic analyses were used to explore the correlations of image manifestations of organ damage and clinical characteristics of heat stroke.Results Among 46 cases,26 were classified as positive group A and 20 as negative group.There were significant differences of patients' age,gender,Glasgow coma scale(GCS)and heart rate at admission,thrombin time,D-dimer,lactate dehydrogenase,partial pressure of carbon dioxide,lactic acid,myoglobin,platelet count(PLC),interleukin-6 and serum creatinine before treatment,as well as time of hospital stay and post-treatment outcomes between groups(all P＜0.05).GCS score,PLC levels and time of hospital stay of heat stroke patients were all correlated with image manifeations of organ damage(OR=0.592,0.729,1.532,all P＜0.05),among which GCS score and time of hospital stay were independently associated with image manifeations of organ damage(OR=0.607,1.448,both P＜0.05).Conclusion Image manifeations of organ damage was correlated with GCS score and time of hospital stay in patients with heat stroke.

Objective:To develop CNN-Dinov2, a deep learning-based automatic classification model for Gram-stained images, enabling rapid diagnosis of three prevalent Gram-negative bacilli in bloodstream infections: Escherichia coli ( E.coli), Klebsiella pneumoniae ( K.pneumoniae), and Pseudomonas aeruginosa ( P.aeruginosa). Methods:This evaluation study analyzed 1 425 Gram-stained microscopic images from patients with bloodstream infections at Houjie Hospital, in Dongguan City, collected between January 2023 and January 2024. The images, all positive for blood culture and identified as target strains, were categorized into Escherichia coli (419 images), Klebsiella pneumoniae (411 images), Pseudomonas aeruginosa (413 images), and other Gram-negative bacilli (182 images). They were randomly split into a training set (1 141 images), a validation set (141 images), and a test set (143 images) in an 8∶1∶1 ratio. A hybrid CNN-Dinov2 model was developed by integrating ResNet′s local feature extraction with Dinov2′s global pre-trained features, followed by a fully connected layer. The model was optimized by inputting the preprocessed images and adjusting parameters through loss calculation and backpropagation. AlexNet, Dinov2, and ResNet18 served as control models. The models′ classification performance was assessed using accuracy, precision, weighted F1 score, and recall rate, derived from the confusion matrix. The PR curve and AP value further evaluated each model′s classification capability across the four image categories. Results:The CNN-Dinov2 model achieved a training accuracy of 99.74%, a validation accuracy of 98.12%, and a validation loss of 0.070 6, demonstrating robust generalization without overfitting. Validation metrics revealed superior performance with an accuracy of 98.60%, precision of 98.65%, a weighted F1 score of 98.60%, and a recall rate of 98.60%, outperforming other models. The confusion matrix confirmed its strong classification capability, with the highest sum of diagonal values for identifying four types of bacteria. The macro average precision (AP) values under the precision-recall (PR) curves were all 1, indicating excellent discrimination across all categories. Overall, the CNN-Dinov2 model exhibited the best performance among the four models evaluated.Conclusion:This study successfully developed CNN-Dinov2, an automated classification model for Gram staining images. It offers valuable support for the rapid diagnosis of bloodstream infections caused by Escherichia coli, Klebsiella pneumoniae, and Pseudomonas aeruginosa, demonstrating practical utility.

Purpose To investigate the clinicopathological features,molecular genetics and prognosis of high grade B cell lymphoma with concurrent MYC rearrangement and 11q aberra-tions(HGBCL-MYC-11q).MethodsThree cases of HGBCL-MYC-11q were reviewed and analyzed using hematoxylin-eosin staining,immunohistochemistry,EBER in situ hybridization and fluorescence in situ hybridization.Clinical data were collected with follow-up.Results All three patients were male,age was 10,61,and 74 years,respectively.All patients had Ann Arbor stage Ⅳ disease.All three cases were biopsies occurring in the nasopharynx,upper pharynx and ileocecus,respectively.Three cases were morphologically similar to diffuse infiltrative growth of tumor cells,moderate or moderately large cells,round to slightly irregular nuclei and easily visible mitotic figures.Focal necrosis was noted in one case.One case exhibited the distinct"starry sky"pattern.All cases expressed CD20,BCL6 and MUM1 and high Ki67 index,two cases expressed CD10 and two cases ex-pressed BCL2.CD3,CD30 and TDT were all negative.EBER in situ hybridization was all negative.FISH analyses using C-MYC break-apart probes were all positive and all cases had 11q aberrations.One case only had the 11q23.3 amplification;and one case only had the 11q24.3 loss.After a follow-up for 1-18 months,one patient died and two patients survived with disease.ConclusionHGBCL-MYC-11q is rare,morphologically similar to BL/HGBCL,with MYC rearrangement and 11q abnormali-ties.We should enhance awareness of the disease and improve more accurate diagnosis and differential diagnosis of the disease.

To investigate the genomic features and perform cluster analysis of Carbapenem-resistant Klebsiella pneumoniae (CRKP) to provide an experimental basis for guiding the prevention and treatment of CRKP infections.A retrospective case-cohort study was conducted on 19 non-redundant CRKP strains isolated from the Tenth Affiliated Hospital of Southern Medical University between January and June 2023. Whole genome sequencing (WGS) and multilocus sequence typing (MLST) were performed to compare genomic features and analyze the resistance genes and homology of the strains.The results showed that the 19 CRKP strains were isolated from 8 different clinical departments, mainly from respiratory specimens. The whole genome sequencing revealed that the genomic lengths of CRKP ranged from 4.90 to 5.85 Mbp, with contigs N50 values>20 kb for each genome. The median overall GC content was 57.0% (50.4%-57.1%). Comparative genomic analysis identified three regions with high genomic variability. WGS detected 32 resistance genes across 11 categories. All 19 strains carried carbapenem resistance genes ( blaKPC-2 and blaOXA-48), blaTEM-1B extended-spectrum β-lactamase resistance genes, qnrS1 quinolone resistance gene, and fosA fosfomycin resistance gene, with each strain carrying only one carbapenemase gene. The detection rate of blaKPC-2 was 94.7% (18/19). MLST identified three sequence types: ST11, ST437 and ST147, with ST11 being predominant (89.5%, 17/19). Clustering analysis based on acquired resistance genes revealed three clonal transmission patterns among strains 72 and 90, and strains 88, 84, 66 and 79.In conclusion, CRKP strains carry multiple resistance genes, and clustering analysis indicating that nosocomial clonal transmission is closely related to acquired resistance genes. The ST11- blaKPC-2 type strain is the predominant clone. Strengthened surveillance and effective control strategies are necessary to reduce nosocomial transmission of CRKP.

Purpose To investigate the clinicopathological features,diagnosis,differential diagnosis and molecular charac-teristics desmoid fibromatosis the breast(DFB).Methods The clinicopathological data and prognostic information of 26 pa-tients with desmoid fibromatosis of the breast were collected.Their clinical characteristics,histological morphologies,immu-nophenotypes and molecular characteristics were analyzed.Re-sults All the 26 patients were female,with a median age of 34.5 years and an average age of 36.8 years(range from 13 to 69 years).There were 10 cases in left breast,14 cases in right breast and 2 cases in bilateral breast.Isolated and painless breast masses were found in all patients,and 3 cases were ac-companied by nipple depression.Grossly,most of them showed a poorly circumscribed tan-white to gray mass with a trabeculated appearance.Microscopically,all the tumors were composed of mild spindle cells with varying intervals of collagen fiber bun-dles.The boundary between the tumor and the surrounding breast tissue was not clear,and the tumor cells infiltrated adja-cent tissues,such as breast ducts,lobules,adipose tissue,and skeletal muscle.Uniform spindle or oval cells formed in fascicles and interwoven,without atypical or polymorphism;the nuclear chromatin was sparse or vacuolar,with small nucleoli,but mito-ses were rare or absent.Nuclear accumulation of β-catenin was present in 20 of 26 tumors,SMA was variously positive in 20 of 26,desmin was focally positive in 6 of 26;CKpan,CK5/6,p63,CD34,CD10 and S-100 were all negative.The Ki67 index was 5％-10％.The gene mutations of CTNNB1 exon 3 were found in 18 of 26 by Sanger sequencing,including T41A(83.3％),S45P(11.1％)and S45F(5.6％).Two patients also had familial adenomatous polyposis.Local resection was performed in 23 cases,mastectomy was performed in 2 cases,and one case did not require any additional treatment after core needle biopsy.20 cases were followed up for 1-108 months,and no recurrence occurred.Conclusion Desmoid fibromatosis of the breast is rare,and can mimic breast cancer clinically,ra-diologically and histologically.It should be always considered in differential diagnosis for the spindle cell proliferations of the breast.A diagnosis of DFB can be achieved basing on the typi-cal histopathology,immunohistochemistry,history and gene se-quencing.

To investigate the genomic features and perform cluster analysis of Carbapenem-resistant Klebsiella pneumoniae (CRKP) to provide an experimental basis for guiding the prevention and treatment of CRKP infections.A retrospective case-cohort study was conducted on 19 non-redundant CRKP strains isolated from the Tenth Affiliated Hospital of Southern Medical University between January and June 2023. Whole genome sequencing (WGS) and multilocus sequence typing (MLST) were performed to compare genomic features and analyze the resistance genes and homology of the strains.The results showed that the 19 CRKP strains were isolated from 8 different clinical departments, mainly from respiratory specimens. The whole genome sequencing revealed that the genomic lengths of CRKP ranged from 4.90 to 5.85 Mbp, with contigs N50 values>20 kb for each genome. The median overall GC content was 57.0% (50.4%-57.1%). Comparative genomic analysis identified three regions with high genomic variability. WGS detected 32 resistance genes across 11 categories. All 19 strains carried carbapenem resistance genes ( blaKPC-2 and blaOXA-48), blaTEM-1B extended-spectrum β-lactamase resistance genes, qnrS1 quinolone resistance gene, and fosA fosfomycin resistance gene, with each strain carrying only one carbapenemase gene. The detection rate of blaKPC-2 was 94.7% (18/19). MLST identified three sequence types: ST11, ST437 and ST147, with ST11 being predominant (89.5%, 17/19). Clustering analysis based on acquired resistance genes revealed three clonal transmission patterns among strains 72 and 90, and strains 88, 84, 66 and 79.In conclusion, CRKP strains carry multiple resistance genes, and clustering analysis indicating that nosocomial clonal transmission is closely related to acquired resistance genes. The ST11- blaKPC-2 type strain is the predominant clone. Strengthened surveillance and effective control strategies are necessary to reduce nosocomial transmission of CRKP.