RESUMO
Semantic segmentation for diagnosing chest-related diseases like cardiomegaly, emphysema, pleural effusions, and pneumothorax is a critical yet understudied tool for identifying the chest anatomy. A dangerous disease among these is cardiomegaly, in which sudden death is a high risk. An expert medical practitioner can diagnose cardiomegaly early using a chest radiograph (CXR). Cardiomegaly is a heart enlargement disease that can be analyzed by calculating the transverse cardiac diameter (TCD) and the cardiothoracic ratio (CTR). However, the manual estimation of CTR and other chest-related diseases requires much time from medical experts. Based on their anatomical semantics, artificial intelligence estimates cardiomegaly and related diseases by segmenting CXRs. Unfortunately, due to poor-quality images and variations in intensity, the automatic segmentation of the lungs and heart with CXRs is challenging. Deep learning-based methods are being used to identify the chest anatomy segmentation, but most of them only consider the lung segmentation, requiring a great deal of training. This work is based on a multiclass concatenation-based automatic semantic segmentation network, CardioNet, that was explicitly designed to perform fine segmentation using fewer parameters than a conventional deep learning scheme. Furthermore, the semantic segmentation of other chest-related diseases is diagnosed using CardioNet. CardioNet is evaluated using the JSRT dataset (Japanese Society of Radiological Technology). The JSRT dataset is publicly available and contains multiclass segmentation of the heart, lungs, and clavicle bones. In addition, our study examined lung segmentation using another publicly available dataset, Montgomery County (MC). The experimental results of the proposed CardioNet model achieved acceptable accuracy and competitive results across all datasets.
RESUMO
Background: Metastasis of the lymph node is one of the most significant prognostic factors for breast cancer (BC). Aim: To predict positivity of the lymph node in BC patients with help of USG and USG-guided FNAC and thus to prevent unnecessary morbidity. Methods: 50 patients of incisional/true cut biopsy-proven BC patients were included. All were subjected to mammography, USG and FNAC of the lump breast. USG-guided FNAC of the axillary lymph node was done in 25 of these patients. These findings were assessed by histological examination following dissection of the axillary lymph node. Results: Axillary lymph node (ALN) metastasis was present in 42 patients on histopathology; 21 patients suspicious of malignancy on preoperative USG were confirmed by HPE. Out of 88 confirmed lymph nodes evaluated on ultrasonography, 4 were benign, 18 were indeterminate and 66 were suspicious. The most promising features were tumour length/depth ratio of <1.5 in 81, absent fatty hilum in 73% and hypoechoic cortex in 74%. Assessment of axilla with USG had a sensitivity of 50%, a specificity of 100%, a PPV of 100%, an NPV of 27.59% and a diagnostic accuracy of 58%. Preoperative USG-guided FNAC had a sensitivity of 91.67%, a specificity of 100%, a PPV of 100%, an NPV of 33.33% and a diagnostic accuracy of 92%. Conclusion: USG can detect non-palpable axillary lymph nodes and FNAC can increase the sensitivity and specificity of this technique, which makes this procedure very promising in detecting axillary metastases in BC patients.
