There is a growing interest in research aimed at better understanding the disease status or predicting the prognosis of patients with simple blood tests associated with systemic inflammation. The neutrophil-lymphocyte ratio (NLR), lymphocyte-monocyte ratio (LMR), platelet-lymphocyte ratio (PLR), and mean platelet volume (MPV) can be used as factors to determine the prognosis of patients in various clinical situations. However, reference values for these attributes based on large, healthy populations have yet to be determined.From January 2014 to December 2016, data from routine blood analyses were collected from healthy patients in the checkup center of a tertiary hospital in Seoul, South Korea. Retrospective data review was then performed on an electronic medical record system. Data were treated anonymously as only age, sex, body mass index, medical history including cancer diagnosis, medications, and smoking status were considered. After the initial screen, we had a collection of 12,160 samples from patients without any medical history, including cancer treatment. This patient pool consisted of 6268 (51.5%, median age 47 years) and 5892 (48.5%, median age 46 years) male and female patients, respectively. The mean NLR across all ages was 1.65 (0.79), and the values for men and women were 1.63 (0.76) and 1.66 (0.82), respectively. The mean LMR, PLR, and MPV were 5.31 (1.68), 132.40 (43.68), and 10.02 (0.79), respectively. This study provides preliminary reference data on LMR, PLR, and MPV from different age and sex groups in South Korea. The results suggest that different cutoff values should be applied to the various patient populations.

No abstract

In early gastric cancer (EGC), tumor invasion depth is an important factor for determining the treatment method. However, as endoscopic ultrasonography has limitations when measuring the exact depth in a clinical setting as endoscopists often depend on gross findings and personal experience. The present study aimed to develop a model optimized for EGC detection and depth prediction, and we investigated factors affecting artificial intelligence (AI) diagnosis. We employed a visual geometry group(VGG)-16 model for the classification of endoscopic images as EGC (T1a or T1b) or non-EGC. To induce the model to activate EGC regions during training, we proposed a novel loss function that simultaneously measured classification and localization errors. We experimented with 11,539 endoscopic images (896 T1a-EGC, 809 T1b-EGC, and 9834 non-EGC). The areas under the curves of receiver operating characteristic curves for EGC detection and depth prediction were 0.981 and 0.851, respectively. Among the factors affecting AI prediction of tumor depth, only histologic differentiation was significantly associated, where undifferentiated-type histology exhibited a lower AI accuracy. Thus, the lesion-based model is an appropriate training method for AI in EGC. However, further improvements and validation are required, especially for undifferentiated-type histology.