Browsing by Author "Moreira, C"
Now showing 1 - 4 of 4
Results Per Page
Sort Options
- Caso radiológicoPublication . Nascimento, J; Gomes, M; Moreira, C; Macedo, F
- Comparing Visual Search Patterns in Chest X-Ray DiagnosticsPublication . Moreira, C; Alvito, DM; Sousa, SC; Nobre, IM; Ouyang, C; Kopper, R; Duchowski, A; Jorge, J
- MDF-Net for abnormality detection by fusing X-rays with clinical dataPublication . Hsieh, C; Nobre, IB; Sousa, SC; Ouyang, C; Brereton, M; Nascimento, JC; Jorge, J; Moreira, CThis study investigates the effects of including patients' clinical information on the performance of deep learning (DL) classifiers for disease location in chest X-ray images. Although current classifiers achieve high performance using chest X-ray images alone, consultations with practicing radiologists indicate that clinical data is highly informative and essential for interpreting medical images and making proper diagnoses. In this work, we propose a novel architecture consisting of two fusion methods that enable the model to simultaneously process patients' clinical data (structured data) and chest X-rays (image data). Since these data modalities are in different dimensional spaces, we propose a spatial arrangement strategy, spatialization, to facilitate the multimodal learning process in a Mask R-CNN model. We performed an extensive experimental evaluation using MIMIC-Eye, a dataset comprising different modalities: MIMIC-CXR (chest X-ray images), MIMIC IV-ED (patients' clinical data), and REFLACX (annotations of disease locations in chest X-rays). Results show that incorporating patients' clinical data in a DL model together with the proposed fusion methods improves the disease localization in chest X-rays by 12% in terms of Average Precision compared to a standard Mask R-CNN using chest X-rays alone. Further ablation studies also emphasize the importance of multimodal DL architectures and the incorporation of patients' clinical data in disease localization. In the interest of fostering scientific reproducibility, the architecture proposed within this investigation has been made publicly accessible( https://github.com/ChihchengHsieh/multimodal-abnormalities-detection ).
- Validation of control of allergic rhinitis and asthma test for children (CARATKids)--a prospective multicenter studyPublication . Linhares, DV; Fonseca, JA; Borrego, LM; Matos, A; Pereira, AM; Sá-Sousa, A; Gaspar, A; Mendes, C; Moreira, C; Gomes, E; Rebelo, FF; Cidrais-Rodrigues, JC; Onofre, JM; Azevedo, LF; Alfaro, M; Calix, MJ; Amaral, R; Rodrigues-Alves, R; Correia-Sousa, J; Morais-Almeida, M; CARATKids study groupBACKGROUND: Control of Allergic Rhinitis and Asthma Test for Children (CARATKids) is the first questionnaire that assesses simultaneously allergic rhinitis and asthma control in children. It was recently developed, but redundancy of questions and its psychometric properties were not assessed. This study aimed to (i) establish the final version of the CARATKids questionnaire and (ii) evaluate its reliability, responsiveness, cross-sectional validity, and longitudinal validity. METHODS: A prospective observational study was conducted in 11 Portuguese centers. During two visits separated by 6 wk, CARATKids, visual analog scale scales and childhood asthma control test were completed, and participant's asthma and rhinitis were evaluated by his/her physician without knowing the questionnaires' results. Data-driven item reduction was conducted, and internal consistency, responsiveness analysis, and associations with external measures of disease status were assessed. RESULTS: Of the 113 children included, 101 completed both visits. After item reduction, the final version of the questionnaire has 13 items, eight to be answered by the child and five by the caregiver. Its Cronbach's alpha was 0.80, the Guyatt's responsiveness index was -1.51, and a significant (p < 0.001) within-patient change of CARATKids score in clinical unstable patients was observed. Regarding cross-sectional validity, correlation coefficients of CARATKids with the external measures of control were between 0.45 and -0.69 and met the a priori predictions. In the longitudinal validity assessment, the correlation coefficients between the score changes of CARATKids and those of external measures of control ranged from 0.34 to 0.46.