D3EGFR: a webserver for deep learning-guided drug sensitivity prediction and drug response information retrieval for EGFR mutation-driven lung cancer.

As key oncogenic drivers in non-small-cell lung cancer (NSCLC), various mutations in the epidermal growth factor receptor (EGFR) with variable drug sensitivities have been a major obstacle for precision medicine. To achieve clinical-level drug recommendations, a platform for clinical patient case retrieval and reliable drug sensitivity prediction is highly expected. Therefore, we built a database, D3EGFRdb, with the clinicopathologic characteristics and drug responses of 1339 patients with EGFR mutations via literature mining. On the basis of D3EGFRdb, we developed a deep learning-based prediction model, D3EGFRAI, for drug sensitivity prediction of new EGFR mutation-driven NSCLC. Model validations of D3EGFRAI showed a prediction accuracy of 0.81 and 0.85 for patients from D3EGFRdb and our hospitals, respectively. Furthermore, mutation scanning of the crucial residues inside drug-binding pockets, which may occur in the future, was performed to explore their drug sensitivity changes. D3EGFR is the first platform to achieve clinical-level drug response prediction of all approved small molecule drugs for EGFR mutation-driven lung cancer and is freely accessible at https://www.d3pharma.com/D3EGFR/index.php.

Allelic phenotype prediction of phenylketonuria based on the machine learning method.

BACKGROUND: Phenylketonuria (PKU) is caused by mutations in the phenylalanine hydroxylase (PAH) gene. Our study aimed to predict the phenotype using the allelic genotype. METHODS: A total of 1291 PKU patients with 623 various variants were used as the training dataset for predicting allelic phenotypes. We designed a common machine learning framework to predict allelic genotypes associated with the phenotype. RESULTS: We identified 235 different mutations and 623 various allelic genotypes. The features extracted from the structure of mutations and graph properties of the PKU network to predict the phenotype of PKU were named PPML (PKU phenotype predicted by machine learning). The phenotype of PKU was classified into three different categories: classical PKU (cPKU), mild PKU (mPKU) and mild hyperphenylalaninemia (MHP). Three hub nodes (c.728G>A for cPKU, c.721 for mPKU and c.158G>A for HPA) were used as each classification center, and 5 node attributes were extracted from the network graph for machine learning training features. The area under the ROC curve was AUC = 0.832 for cPKU, AUC = 0.678 for mPKU and AUC = 0.874 for MHP. This suggests that PPML is a powerful method to predict allelic phenotypes in PKU and can be used for genetic counseling of PKU families. CONCLUSIONS: The web version of PPML predicts PKU allele classification supported by applicable real cases and prediction results. It is an online database that can be used for PKU phenotype prediction http://www.bioinfogenetics.info/PPML/ .

Application of the process-based teaching based on SPARK case database in the practice teaching of radiology in the musculoskeletal system for undergraduate medical students.

BACKGROUND: Process-based teaching is a new education model. SPARK case database is a free medical imaging case database. This manuscript aimed to explore the application of the process-based teaching based on SPARK case database in the practice teaching of radiology in the musculoskeletal system. METHODS: 117 third year medical students were included. They were divided into Group A, B, C and D according to the curriculum arrangement. Group A and B attended the experimental class at the same time, A was the experimental group, B was the control group. Group C and D attended experimental classes at the same time, C was the experimental group, D was the control group. The experimental group used SPARK case database, while the control group used traditional teaching model for learning. The four groups of students were respectively tested after the theoretical class, before the experimental class, after the experimental class, and one week after the experimental class to compare the results. Finally, all students used SPARK case database to study, and were tested one month after the experimental class to compare their differences. RESULTS: The scores after the theoretical class of Group A and B were (100.0 ± 25.4), (101.0 ± 23.8)(t=-0.160, P > 0.05), Group C and D were (94.7 ± 23.7), (92.1 ± 18.6)(t = 0.467, P > 0.05). The scores of Group A and B before and after the experimental class and one week after the experimental class were respectively (84.1 ± 17.4), (72.1 ± 21.3)(t = 2.363, P < 0.05), (107.6 ± 14.3), (102.1 ± 18.0)(t = 1.292, P > 0.05), (89.7 ± 24.3), (66.6 ± 23.2)(t = 3.706, P < 0.05). The scores of Group C and D were (94.0 ± 17.3), (72.8 ± 25.5)(t = 3.755, P < 0.05), (107.3 ± 20.3), (93.1 ± 20.9)(t = 2.652, P < 0.05), (100.3 ± 19.7), (77.2 ± 24.0)(t = 4.039, P < 0.05). The scores of Group A and B for one month after the experimental class were (86.6 ± 28.8), (84.5 ± 24.0)(t = 0.297, P > 0.05), and Group C and D were (95.7 ± 20.3), (91.7 ± 23.0)(t = 0.699, P > 0.05). CONCLUSIONS: The process-based teaching based on SPARK case database could improve the radiology practice ability of the musculoskeletal system of students.

Finally Throwing Those Wellies Away? Collaborating in Search of a Solution for Venice Flooding.

Collaborative governance is often advocated as a way to address 'messy' problems that individual stakeholders cannot solve alone. However, whereas stakeholders' participation brings a broad range of response options to public decision-making, the complexities of the perspectives at stake may also lead to conflicts and stalemates. This is especially true in collaborative environmental governance, where conflict is common and stakeholders' interdependence in more than one arena tends to be frequent. Based on a longitudinal field study, we explore how to break stalemates in collaborative environmental governance when they occur, and move the collaboration towards a shared decision. The successful collaborative decision-making for the defence of Venice against floods represents our empirical setting. Our findings show that, in this context, the combined effect of three factors seems to be important to break stalemates and lead stakeholders towards a shared decision in collaborative environmental governance: stakeholders' reactivation, fear of marginalization and leaders acting as orchestrators.

Technical note: A report on the Forensic Anthropology Database for Assessing Methods Accuracy.

We report on the functionality, available support, and research capability of the Forensic Anthropology Database for Assessing Methods Accuracy (FADAMA; DOJ DUBX0213). FADAMA is an online repository for case data from identified forensic skeletal cases. The goal of FADAMA is to address the lack of adequate measures for assessing accuracy and reliability of forensic anthropology methods. FADAMA requires users to apply for access with their university or organization credentials. Verified users may upload and download anonymized case data via the user interface, after signing a terms of service agreement outlining ethical behavior. Case data uploads require information about the actual biological profile of the decedent and the forensic anthropology estimations. Uploading case data takes approximately 15-25 min. FADAMA users currently have 85 methods to select from when entering case data, with the capability to add new methods as they are developed. Access to the database is free, and online video tutorials are available for users covering database functionality. Currently, the database houses anonymized case data for over 350 identified cases from across the U.S. Funding has been allocated for a database technician to assist offices with large caseloads to upload cases. As it stands, the database is easy to use, and maintains thoughtful tools to assist users. The power of the database to identify trends in both method accuracy and usage is apparent, and will continue to grow as more cases are added.

Development of a CT image case database and content-based image retrieval system for non-cancerous respiratory diseases: Method and preliminary assessment.

BACKGROUND: To evaluate the benefits of using a CT image case database (DB) with content-based image retrieval system for the diagnosis of typical non-cancerous respiratory diseases. METHODS: Using this DB, which comprised data on 191 cases covering 69 diseases, 933 imaging findings that contributed to differential diagnoses were annotated. Ten test cases were selected. Image similarity between each marked test case lesion and the lesions of the top 10 retrieved cases were assessed and classified as similar, somewhat similar, or dissimilar by two physicians in consensus. Additionally, the accuracy of five internal medicine residents' abilities to interpret CT findings and provide disease diagnoses with and without the proposed system was evaluated by image interpretation experiments involving five test cases. The rates of concordance between the subjects' interpretations and the correct answers prepared in advance by two specialists in consensus were converted into scores. RESULTS: The mean (± SD) of image similarity among the 10 test cases was as follows: 5.1 ± 2.7 (similar), 2.9 ± 1.0 (somewhat similar), and 2.0 ± 2.4 (dissimilar). Using the proposed system, the subjects' mean score for the correct interpretation of CT findings improved from 15.1 to 28.2 points (p = 0.131) and for the correct disease diagnoses, from 9.3 to 28.2 points (p = 0.034). CONCLUSIONS: Although this was a preliminary small-scale assessment, the results suggest that this system may contribute to an improved interpretation of CT findings and differential diagnosis of non-cancerous respiratory diseases, which are difficult to diagnose for inexperienced physicians.