Deep convolutional neural network algorithm for the automatic segmentation of oral potentially malignant disorders and oral cancers.

This study aimed to develop an algorithm to automatically segment the oral potentially malignant diseases (OPMDs) and oral cancers (OCs) of all oral subsites with various deep convolutional neural network applications. A total of 510 intraoral images of OPMDs and OCs were collected over 3 years (2006-2009). All images were confirmed both with patient records and histopathological reports. Following the labeling of the lesions the dataset was arbitrarily split, using random sampling in Python as the study dataset, validation dataset, and test dataset. Pixels were classified as the OPMDs and OCs with the OPMD/OC label and the rest as the background. U-Net architecture was used and the model with the best validation loss was chosen for the testing among the trained 500 epochs. Dice similarity coefficient (DSC) score was noted. The intra-observer ICC was found to be 0.994 while the inter-observer reliability was 0.989. The calculated DSC and validation accuracy across all clinical images were 0.697 and 0.805, respectively. Our algorithm did not maintain an excellent DSC due to multiple reasons for the detection of both OC and OPMDs in oral cavity sites. A better standardization for both 2D and 3D imaging (such as patient positioning) and a bigger dataset are required to improve the quality of such studies. This is the first study which aimed to segment OPMDs and OCs in all subsites of oral cavity which is crucial not only for the early diagnosis but also for higher survival rates.

MDS-associated somatic mutations and clonal hematopoiesis are common in idiopathic cytopenias of undetermined significance.

Establishing a diagnosis in patients suspected of having a myelodysplastic syndrome (MDS) can be challenging and could be informed by the identification of somatic mutations. We performed a prospective study to examine the frequency and types of mutations encountered in 144 patients with unexplained cytopenias. Based on bone marrow findings, 17% were diagnosed with MDS, 15% with idiopathic cytopenias of undetermined significance (ICUS) and some evidence of dysplasia, and 69% with ICUS and no dysplasia. Bone marrow DNA was sequenced for mutations in 22 frequently mutated myeloid malignancy genes. Somatic mutations were identified in 71% of MDS patients, 62% of patients with ICUS and some dysplasia, and 20% of ICUS patients and no dysplasia. In total, 35% of ICUS patients carried a somatic mutation or chromosomal abnormality indicative of clonal hematopoiesis. We validated these results in a cohort of 91 lower-risk MDS and 249 ICUS cases identified over a 6-month interval. Mutations were found in 79% of those with MDS, in 45% of those with ICUS with dysplasia, and in 17% of those with ICUS without dysplasia. The spectrum of mutated genes was similar with the exception of SF3B1 which was rarely mutated in patients without dysplasia. Variant allele fractions were comparable between clonal ICUS (CCUS) and MDS as were mean age and blood counts. We demonstrate that CCUS is a more frequent diagnosis than MDS in cytopenic patients. Clinical and mutational features are similar in these groups and may have diagnostic utility once outcomes in CCUS patients are better understood.