METHODS: In this study, we propose a multi-view secondary input residual (MV-SIR) convolutional neural network model for 3D lung nodule segmentation using the Lung Image Database Consortium and Image Database Resource Initiative (LIDC-IDRI) dataset of chest computed tomography (CT) images. Lung nodule cubes are prepared from the sample CT images. Further, from the axial, coronal, and sagittal perspectives, multi-view patches are generated with randomly selected voxels in the lung nodule cubes as centers. Our model consists of six submodels, which enable learning of 3D lung nodules sliced into three views of features; each submodel extracts voxel heterogeneity and shape heterogeneity features. We convert the segmentation of 3D lung nodules into voxel classification by inputting the multi-view patches into the model and determine whether the voxel points belong to the nodule. The structure of the secondary input residual submodel comprises a residual block followed by a secondary input module. We integrate the six submodels to classify whether voxel points belong to nodules, and then reconstruct the segmentation image.
RESULTS: The results of tests conducted using our model and comparison with other existing CNN models indicate that the MV-SIR model achieves excellent results in the 3D segmentation of pulmonary nodules, with a Dice coefficient of 0.926 and an average surface distance of 0.072.
CONCLUSION: our MV-SIR model can accurately perform 3D segmentation of lung nodules with the same segmentation accuracy as the U-net model.
METHODS: An online/face-to-face, questionnaire-based survey of respiratory specialists and primary care physicians from eight Asian countries/region was carried out. The survey explored asthma control, inhaler selection, technique and use; physician-patient communications and asthma education. Inclusion criteria were >50% of practice time spent on direct patient care; and treated >30 patients with asthma per month, of which >60% were aged >12 years.
RESULTS: REALISE Asia (Phase 2) involved 375 physicians with average 15.9(±6.8) years of clinical experience. 89.1% of physicians reporting use of guidelines estimated that 53.2% of their patients have well-controlled (GINA-defined) asthma. Top consideration for inhaler choice was asthma severity (82.4%) and lowest, socio-economic status (32.5%). Then 54.7% of physicians checked their patients' inhaler techniques during consultations but 28.2(±19.1)% of patients were using their inhalers incorrectly; 21.1-57.9% of physicians could spot improper inhaler techniques in video demonstrations. And 79.6% of physicians believed combination inhalers could increase adherence because of convenience (53.7%), efficacy (52.7%) and usability (18.9%). Initial and follow-up consultations took 16.8(±8.4) and 9.2(±5.3) minutes, respectively. Most (85.1%) physicians used verbal conversations and least (24.5%), video demonstrations of inhaler use; 56.8% agreed that patient attitudes influenced their treatment approach.
CONCLUSION: Physicians and patients have different views of 'well-controlled' asthma. Although physicians informed patients about asthma and inhaler usage, they overestimated actual usage and patients' knowledge was sub-optimal. Physician-patient interactions can be augmented with understanding of patient attitudes, visual aids and ancillary support to perform physical demonstrations to improve treatment outcomes.