Search

Abstract Background Breast cancer is the most common malignant tumor in the world. Intraoperative frozen section of sentinel lymph nodes is an important basis for determining whether axillary lymph node dissection is required for breast cancer surgery. We propose an RRCART model based on a deep-learning network to identify metastases in 2362 frozen sections and count the wrongly identified sections and the associated reasons. The purpose is to summarize the factors that affect the accuracy of the artificial intelligence model and propose corresponding solutions. Methods We took the pathological diagnosis of senior pathologists as the gold standard and identified errors. The pathologists and artificial intelligence engineers jointly read the images and heatmaps to determine the locations of the identified errors on sections, and the pathologists found the reasons (false reasons) for the errors. Through NVivo 12 Plus, qualitative analysis of word frequency analysis and nodal analysis was performed on the error reasons, and the top-down error reason framework of “artificial intelligence RRCART model to identify frozen sections of breast cancer lymph nodes” was constructed based on the importance of false reasons. Results There were 101 incorrectly identified sections in 2362 slides, including 42 false negatives and 59 false positives. Through NVivo 12 Plus software, the error causes were node-coded, and finally, 2 parent nodes (high-frequency error, low-frequency error) and 5 child nodes (section quality, normal lymph node structure, secondary reaction of lymph nodes, micrometastasis, and special growth pattern of tumor) were obtained; among them, the error of highest frequency was that caused by normal lymph node structure, with a total of 45 cases (44.55%), followed by micrometastasis, which occurred in 30 cases (29.70%). Conclusions The causes of identification errors in examination of sentinel lymph node frozen sections by artificial intelligence are, in descending order of influence, normal lymph node structure, micrometastases, section quality, special tumor growth patterns and secondary lymph node reactions. In this study, by constructing an artificial intelligence model to identify the error causes of frozen sections of lymph nodes in breast cancer and by analyzing the model in detail, we found that poor quality of slices was the preproblem of many identification errors, which can lead to other errors, such as unclear recognition of lymph node structure by computer. Therefore, we believe that the process of artificial intelligence pathological diagnosis should be optimized, and the quality control of the pathological sections included in the artificial intelligence reading should be carried out first to exclude the influence of poor section quality on the computer model. For cases of micrometastasis, we suggest that by differentiating slices into high- and low-confidence groups, low-confidence micrometastatic slices can be separated for manual identification. The normal lymph node structure can be improved by adding samples and training the model in a targeted manner.

Abstract Background Sorbus sensu stricto (Sorbus s.s.) is a genus with important economical values because of its beautiful leaves, and flowers and especially the colorful fruits. It belongs to the tribe Maleae of the family Rosaceae, and comprises about 90 species mainly distributed in China. There is on-going dispute about its infrageneric classification and species delimitation as the species are morphologically similar. With the aim of shedding light on the circumscription of taxa within the genus, phylogenetic analyses were performed using 29 Sorbus s.s. chloroplast (cp) genomes (16 newly sequenced) representing two subgenera and eight sections. Results The 16 cp genomes newly sequenced range between 159,646 bp and 160,178 bp in length. All the samples examined and 22 taxa re-annotated in Sorbus sensu lato (Sorbus s.l.) contain 113 unique genes with 19 of these duplicated in the inverted repeat (IR). Six hypervariable regions including trnR-atpA, petN-psbM, rpl32-trnL, trnH-psbA, trnT-trnL and ndhC-trnV were screened and 44–53 SSRs and 14–31 dispersed repeats were identified as potential molecular markers. Phylogenetic analyses under ML/BI indicated that Sorbus s.l. is polyphyletic, but Sorbus s.s. and the other five segregate genera, Aria, Chamaemespilus, Cormus, Micromeles and Torminalis are monophyletic. Two major clades and four sub-clades resolved with full-support within Sorbus s.s. are not consistent with the existing infrageneric classification. Two subgenera, subg. Sorbus and subg. Albocarmesinae are supported as monophyletic when S. tianschanica is transferred to subg. Albocarmesinae from subg. Sorbus and S. hupehensis var. paucijuga transferred to subg. Sorbus from subg. Albocarmesinae, respectively. The current classification at sectional level is not supported by analysis of cp genome phylogeny. Conclusion Phylogenomic analyses of the cp genomes are useful for inferring phylogenetic relationships in Sorbus s.s. Though genome structure is highly conserved in the genus, hypervariable regions and repeat sequences used are the most promising molecule makers for population genetics, species delimitation and phylogenetic studies.

Abstract Background Several models and markers were developed and found to predict outcome of advanced renal cell carcinoma. This study aimed to evaluate the prognostic value of the ratio of maximum to minimum tumor diameter (ROD) in metastatic clear cell renal cell carcinoma (mccRCC). Methods Patients with mccRCC (n = 213) treated with sunitinib from January 2008 to December 2018 were identified. Cutoff value for ROD was determined using receiver operating characteristic. Patients with different ROD scores were grouped and evaluated. Survival outcomes were estimated by Kaplan–Meier method. Results The optimal ROD cutoff value of 1.34 was determined for progression free survival (PFS) and overall survival (OS). Patients in ROD ≥ 1.34 group had shorter PFS (9.6 versus 17.7 months, p

Abstract Background Non-invasive, especially the urine-based diagnosis of prostate cancer (PCa) remains challenging. Although prostate cancer antigen (PSA) is widely used in prostate cancer screening, the false positives may result in unnecessary invasive procedures. PSA elevated patients are triaged to further evaluation of free/total PSA ratio (f/t PSA), to find out potential clinically significant PCa before undergoing invasive procedures. Genomic instability, especially chromosomal copy number variations (CNVs) were proved much more tumor specific. Here we performed a prospective study to evaluate the diagnostic value of CNV via urine-exfoliated cell DNA analysis in PCa. Methods We enrolled 28 PSA elevated patients (≥ 4 ng/ml), including 16 PCa, 9 benign prostate hypertrophy (BPH) and 3 prostatic intraepithelial neoplasia (PIN). Fresh initial portion urine was collected after hospital admission. Urine exfoliated cell DNA was analyzed by low coverage Whole Genome Sequencing, followed by CNV genotyping by the prostate cancer chromosomal aneuploidy detector (ProCAD). CNVs were quantified in absolute z-score (|Z|). Serum free/total PSA ratio (f/t PSA) was reported altogether. Results In patients with PCa, the most frequent CNV events were chr3q gain (n = 2), chr8q gain (n = 2), chr2q loss (n = 4), and chr18q loss (n = 3). CNVs were found in 81.2% (95% Confidence Interval (CI) 53.7–95.0%) PCa. No CNV was identified in BPH patients. A diagnosis model was established by incorporating all CNVs. At the optimal cutoff of |Z|≥ 2.50, the model reached an AUC of 0.91 (95% CI 0.83–0.99), a sensitivity of 81.2% and a specificity of 100%. The CNV approach significantly outperformed f/t PSA (AUC = 0.62, P = 0.012). Further analyses showed that the CNV positive rate was significantly correlated with tumor grade. CNVs were found in 90.9% (95% CI 57.1–99.5%) high grade tumors and 60.0% (95% CI 17.0–92.7%) low grade tumors. No statistical significance was found for patient age, BMI, disease history and family history. Conclusions Urine exfoliated cells harbor enriched CNV features in PCa patients. Urine detection of CNV might be a biomarker for PCa diagnosis, especially in terms of the clinically significant high-grade tumors.

Abstract Background We aimed to build a common terminology in the domain of cervical cancer, named Cervical Cancer Common Terminology (CCCT), that will facilitate clinical data exchange, ensure quality of data and support large scale data analysis. Methods The standard concepts and relations of CCCT were collected from ICD-10-CM Chinese Version, ICD-9-PC Chinese Version, officially issued commonly used Chinese clinical terms, Chinese guidelines for diagnosis and treatment of cervical cancer and Chinese medical book Lin Qiaozhi Gynecologic Oncology. 2062 cervical cancer electronic medical records (EMRs) from 16 hospitals, belong to different regions and hospital tiers, were collected for terminology enrichment and building common terms and relations. Concepts hierarchies, terms and relationships were built using Protégé. The performance of natural language processing results was evaluated by average precision, recall, and F1-score. The usability of CCCT were evaluated by terminology coverage. Results A total of 880 standard concepts, 1182 common terms, 16 relations and 6 attributes were defined in CCCT, which organized in 6 levels and 11 classes. Initial evaluation of the natural language processing results demonstrated average precision, recall, and F1-score percentages of 96%, 72.6%, and 88.5%. The average terminology coverage for three classes of terms, clinical manifestation, treatment, and pathology, were 87.22%, 92.63%, and 89.85%, respectively. Flexible Chinese expressions exist between regions, traditions, cultures, and language habits within the country, linguistic variations in different settings and diverse translation of introduced western language terms are the main reasons of uncovered terms. Conclusions Our study demonstrated the initial results of CCCT construction. This study is an ongoing work, with the update of medical knowledge, more standard clinical concepts will be added in, and with more EMRs to be collected and analyzed, the term coverage will be continuing improved. In the future, CCCT will effectively support clinical data analysis in large scale.

Abstract Background Previous related studies have mainly focused on renal cell carcinoma (RCC) with venous tumor thrombus, specifically inferior vena cava tumor thrombus with renal vein tumor thrombus (RVTT). However, only a few studies have focused on postoperative long-term survival of RCC patients exclusively with RVTT. Our aim was to investigate the independent prognostic factors for locally advanced RCC with RVTT in China. Methods Patients with locally advanced RCC with RVTT were enrolled for the study from January 2000 to December 2015. All patients underwent radical nephrectomy. Survival analysis was estimated using Kaplan-Meier. Univariable and multivariable survival analyses were performed using COX. Patients were divided into high-risk, middle-risk, and low-risk groups based on independent prognostic factors and then analyzed for survival. Results One hundred twenty-eight consecutive patients (103 men & 25 women) were enrolled with a median age of 61 years. Thrombi were all graded 0 using the Mayo system, of which 23 were friable. None of the thrombi detached during surgery. 121 patients were successfully followed up, with a median follow-up period of 47 months. Median overall survival was 127 months (95%CI: 101–153). The 5-year and 10-year cancer-specific survival (CSS) rate was 67.9 and 57.0%. 59 patients had recurrence with median time of 40 months. Friable thrombus, paraneoplastic syndrome (PNS), modified Fuhrman grade 3/4 and perirenal fat invasion were independent prognostic factors (p