RESUMO
Since 2000, there have been more than 8000 publications on radiology artificial intelligence (AI). AI breakthroughs allow complex tasks to be automated and even performed beyond human capabilities. However, the lack of details on the methods and algorithm code undercuts its scientific value. Many science subfields have recently faced a reproducibility crisis, eroding trust in processes and results, and influencing the rise in retractions of scientific papers. For the same reasons, conducting research in deep learning (DL) also requires reproducibility. Although several valuable manuscript checklists for AI in medical imaging exist, they are not focused specifically on reproducibility. In this study, we conducted a systematic review of recently published papers in the field of DL to evaluate if the description of their methodology could allow the reproducibility of their findings. We focused on the Journal of Digital Imaging (JDI), a specialized journal that publishes papers on AI and medical imaging. We used the keyword "Deep Learning" and collected the articles published between January 2020 and January 2022. We screened all the articles and included the ones which reported the development of a DL tool in medical imaging. We extracted the reported details about the dataset, data handling steps, data splitting, model details, and performance metrics of each included article. We found 148 articles. Eighty were included after screening for articles that reported developing a DL model for medical image analysis. Five studies have made their code publicly available, and 35 studies have utilized publicly available datasets. We provided figures to show the ratio and absolute count of reported items from included studies. According to our cross-sectional study, in JDI publications on DL in medical imaging, authors infrequently report the key elements of their study to make it reproducible.
Assuntos
Inteligência Artificial , Diagnóstico por Imagem , Humanos , Estudos Transversais , Reprodutibilidade dos Testes , AlgoritmosRESUMO
The application of deep learning (DL) in medicine introduces transformative tools with the potential to enhance prognosis, diagnosis, and treatment planning. However, ensuring transparent documentation is essential for researchers to enhance reproducibility and refine techniques. Our study addresses the unique challenges presented by DL in medical imaging by developing a comprehensive checklist using the Delphi method to enhance reproducibility and reliability in this dynamic field. We compiled a preliminary checklist based on a comprehensive review of existing checklists and relevant literature. A panel of 11 experts in medical imaging and DL assessed these items using Likert scales, with two survey rounds to refine responses and gauge consensus. We also employed the content validity ratio with a cutoff of 0.59 to determine item face and content validity. Round 1 included a 27-item questionnaire, with 12 items demonstrating high consensus for face and content validity that were then left out of round 2. Round 2 involved refining the checklist, resulting in an additional 17 items. In the last round, 3 items were deemed non-essential or infeasible, while 2 newly suggested items received unanimous agreement for inclusion, resulting in a final 26-item DL model reporting checklist derived from the Delphi process. The 26-item checklist facilitates the reproducible reporting of DL tools and enables scientists to replicate the study's results.
Assuntos
Lista de Checagem , Aprendizado Profundo , Técnica Delphi , Diagnóstico por Imagem , Humanos , Reprodutibilidade dos Testes , Diagnóstico por Imagem/métodos , Diagnóstico por Imagem/normas , Inquéritos e QuestionáriosRESUMO
The significant role played by docking algorithms in drug discovery combined with their serious pitfalls prompted us to envisage a novel concept for validating docking solutions, namely, docking-based comparative intermolecular contacts analysis (dbCICA). This novel approach is based on the number and quality of contacts between docked ligands and amino acid residues within the binding pocket. It assesses a particular docking configuration on the basis of its ability to align a set of ligands within a corresponding binding pocket in such a way that potent ligands come into contact with binding site spots distinct from those approached by low-affinity ligands and vice versa. In other words, dbCICA evaluates the consistency of docking by assessing the correlation between ligands' affinities and their contacts with binding site spots. Optimal dbCICA models can be translated into valid pharmacophore models that can be used as 3-D search queries to mine structural databases for new bioactive compounds. dbCICA was implemented to search for new inhibitors of candida N-myristoyl transferase as potential antifungal agents and glycogen phosphorylase (GP) inhibitors as potential antidiabetic agents. The process culminated in five selective micromolar antifungal leads and nine GP inhibitory leads.
Assuntos
Aciltransferases/antagonistas & inibidores , Inibidores Enzimáticos/farmacologia , Glicogênio Fosforilase/antagonistas & inibidores , Simulação por Computador , Simulação de Acoplamento Molecular , Relação Quantitativa Estrutura-AtividadeRESUMO
BACKGROUND: In this work, support vector regression (SVR) was applied to the optimization of extended release from swellable hydrophilic pentoxifylline matrix-tablets and compared to multiple linear regression (MLR). METHODS: Binary mixtures comprising ethylcellulose and sodium alginate were used as the matrix-former. The matrix-former : drug weight ratio and the percentage of sodium alginate in the matrix-former were the formulation factors (independent variables) and the percentages of drug release at four different time intervals were the responses (dependent variables). Release was determined according to United States Pharmacopeia 31 for 11 pentoxifylline matrix-tablet formulations of different independent variable levels and the corresponding results were used as tutorial data for the construction of an optimized SVR model. Six additional checkpoint matrix-tablet formulations, within the experimental domain, were used to validate the external predictability of SVR and MLR models. RESULTS: It was found that the constructed SVR model fitted better to the release data than the MLR model (higher coefficients of determination, R( 2), lower prediction error sum of squares, narrower range of residuals, and lower mean relative error), outlining its advantages in handling complex nonlinear problems. Superimposed contour plots derived by using the SVR model and describing the effects of polymer and sodium alginate content on pentoxifylline release showed that formulation of optimal release profiles, according to United States Pharmacopeia limitations, could be located at drug : matrix ratio of 1 and sodium alginate content 25% w/w in the matrix-former. CONCLUSION: The results indicate the high potential for SVR in formulation development and Quality by Design.
Assuntos
Química Farmacêutica/métodos , Preparações de Ação Retardada/química , Portadores de Fármacos/química , Pentoxifilina/química , Comprimidos/química , Comprimidos/síntese química , Alginatos/química , Celulose/análogos & derivados , Celulose/química , Ácido Glucurônico/química , Ácidos Hexurônicos/química , Modelos Lineares , Pentoxifilina/administração & dosagem , Análise de RegressãoRESUMO
Background. ß-Thalassemia is a common inherited hemolytic disorder in Palestine. Red blood cell (RBC) transfusion is the principal treatment but it may cause RBC alloimmunization. This study was conducted to determine the prevalence and characteristics of RBC alloimmunization among thalassemic patients in northern governorates of Palestine. Methods. A prospective multicenter observational study was conducted in the thalassemia transfusion centers in the northern governorates of Palestine. The study included 215 thalassemia patients who received regular blood transfusions. Clinical and transfusion records of patients were examined. Antibody screening and identification was conducted using the microcolum gel technique. Results. Two hundred fifteen patients were included in the study. More than half (52.1%) of the patients were males. The median age of patients was 18 years (range: 12-24 years). The most frequent blood group was A (40.5%). Alloantibodies were detected in 12.6% of patients. Anti-D (33.3%), anti-K (25.9%) and anti-E (14.8%) were the most commonly isolated antibodies. There was no association between age, sex, starting age of transfusion, number of transfused units, history of splenectomy and alloimmunization. Conclusions. Anti-Rh and anti-K antibodies were common among this cohort of patients. Age, sex, starting age of transfusion, number of transfused units, and history of splenectomy could not predict the occurrence of alloimmunization.
RESUMO
The pharmacophoric space of streptococcal MurF was explored using a set of 39 known inhibitors. Subsequently, genetic algorithm and multiple linear regression analysis were employed to select an optimal combination of pharmacophoric models and physicochemical descriptors that access self-consistent quantitative structure-activity relationship (QSAR) (r(2)=0.93,F=56.9,r(LOO)(2)=0.91,r(PRESS)(2) against eight external test inhibitors=0.75). Two orthogonal pharmacophores (of cross-correlation r(2)=0.26) emerged in the QSAR equation suggesting the existence of at least two distinct binding modes accessible to ligands within MurF binding pocket. The validity of the QSAR equation and the associated pharmacophore models was experimentally established by the identification of three promising new MurF inhibitors retrieved from the NCI database. Docking studies conducted on active hits supported the binding modes suggested by the pharmacophore/QSAR analysis.