Computer-aided detection (CADe) of pulmonary nodules is critical to assisting radiologists in early identification of lung cancer from computed tomography (CT) scans. This paper proposes a novel CADe system based on a hierarchical vector quantization (VQ) scheme. Compared with the commonly-used simple thresholding approach, high-level VQ yields a more accurate segmentation of the lungs from the chest volume. In identifying initial nodule candidates (INCs) within the lungs, low-level VQ proves to be effective for INCs detection and segmentation, as well as computationally efficient compared to existing approaches. False-positive (FP) reduction is conducted via rule-based filtering operations in combination with a feature-based support vector machine classifier. The proposed system was validated on 205 patient cases from the publically available on-line LIDC (Lung Image Database Consortium) database, with each case having at least one juxta-pleural nodule annotation. Experimental results demonstrated that our CADe system obtained an overall sensitivity of 82.7% at a specificity of 4 FPs/scan, and 89.2% sensitivity at 4.14 FPs/scan for the classification of juxta-pleural INCs only. With respect to comparable CADe systems, the proposed system shows outperformance and demonstrates its potential for fast and adaptive detection of pulmonary nodules via CT imaging.

No abstract

Clinicopathological evidence supports endometrial atypical hyperplasia (AH) or endometrial intraepithelial neoplasia as the precursor of uterine endometrioid carcinoma (EC), the most common gynecologic malignancy. However, the pathogenic progression from AH to EC remains unclear. Here, we employed whole-exome sequencing to identify somatic mutations and copy number changes in micro-dissected lesions from 30 pairs of newly diagnosed AH and EC. We found that all but one pair of AHs shared the same DNA mismatch repair status as their corresponding ECs. The percentage of common mutations between AH lesions and corresponding ECs varied significantly, ranging from 0.1% to 82%. Microsatellite stable AHs had fewer cancer driver mutations than ECs (5 versus 7, p = 0.017), but among microsatellite unstable AHs and ECs there was no difference in mutational numbers (36 versus 38, p = 0.65). As compared to AH specimens, 19 (79%) of 24 microsatellite stable EC tumors gained new cancer driver mutations, most of which involved PTEN, ARID1A, PIK3CA, CTNNB1, or CHD4. Our results suggest that some AH lesions are the immediate precursor of ECs, and progression depends on acquisition of additional cancer driver mutations. However, a complex clonal relationship between AH and EC can also be appreciated, as in some cases both lesions diverge very early or arise independently, thus co-developing with distinct genetic trajectories. Our genome-wide profile of mutations in AH and EC shines new light on the molecular landscape of tumor progression.