Development of a novel artificial intelligence algorithm to detect pulmonary nodules on chest radiography.

BACKGROUND: In this study, we aimed to develop a novel artificial intelligence (AI) algorithm to support pulmonary nodule detection, which will enable physicians to efficiently interpret chest radiographs for lung cancer diagnosis. METHODS: We analyzed chest X-ray images obtained from a health examination center in Fukushima and the National Institutes of Health (NIH) Chest X-ray 14 dataset. We categorized these data into two types: type A included both Fukushima and NIH datasets, and type B included only the Fukushima dataset. We also demonstrated pulmonary nodules in the form of a heatmap display on each chest radiograph and calculated the positive probability score as an index value. RESULTS: Our novel AI algorithms had a receiver operating characteristic (ROC) area under the curve (AUC) of 0.74, a sensitivity of 0.75, and a specificity of 0.60 for the type A dataset. For the type B dataset, the respective values were 0.79, 0.72, and 0.74. The algorithms in both the type A and B datasets were superior to the accuracy of radiologists and similar to previous studies. CONCLUSIONS: The proprietary AI algorithms had a similar accuracy for interpreting chest radiographs when compared with previous studies and radiologists. Especially, we could train a high quality AI algorithm, even with our small type B data set. However, further studies are needed to improve and further validate the accuracy of our AI algorithm.

Molecular and functional characterization of a novel aryl hydrocarbon receptor isoform, AHR1ß, in the chicken (Gallus gallus).

Dioxins including 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) cause toxic effects through activation of the aryl hydrocarbon receptor (AHR)-mediated signaling pathway. Our previous studies have investigated the function of 2 AHR isoforms (AHR1 and AHR2) in avian species and identified a third AHR in the chicken (Gallus gallus) genome. Knowledge of multiple avian AHRs is indispensable to understand molecular mechanisms of AHR-mediated toxic effects and establish risk assessment framework for environmental AHR ligands in avian species. In this study, we successfully isolated a third novel AHR1-like cDNA from chicken and designated it as chicken AHR1 beta (ckAHR1ß). The mRNA expression of ckAHR1ß was primarily detected in the liver, and the hepatic protein expression was confirmed by Western blotting. Although mRNA expression of ckAHR1ß was not altered by in ovo TCDD exposure, ckAHR1ß exhibited specific binding to [(3)H]TCDD, TCDD-dependent nuclear translocation, and interaction with xenobiotic responsive elements (XREs) and AHR nuclear translocators (ARNTs). In vitro XRE-driven reporter gene assays revealed ckAHR1ß-mediated transactivation of TCDD in a dose-dependent manner, showing a 10-fold reduced sensitivity (high EC50) compared with that mediated by ckAHR1. The mutation of Val(371) to Ser(371) in the ligand-binding domain of ckAHR1ß shifted the TCDD-EC50 toward the value observed in ckAHR1, indicating the critical roles of the amino acid in sensitivity. Furthermore, ckAHR1ß-mediated transactivation of TCDD was enhanced by 17ß-estradiol (E2)-activated chicken estrogen receptor α (ckERα), suggesting a positive cross talk between ckERα and ckAHR1ß signaling pathway. Both TCDD-induced and its enhanced activities by E2 were suppressed by the ckAHR repressor in a manner similar to ckAHR1. Collectively, our findings discover the role of ckAHR1ß in dioxin toxicity and give an insight into the evolutionary history of the AHR signaling pathway.

Molecular and functional characterization of aryl hydrocarbon receptor nuclear translocator 1 (ARNT1) and ARNT2 in chicken (Gallus gallus).

Our previous studies have provided evidence that birds have two isoforms of aryl hydrocarbon receptors (AHR1 and AHR2) and AHR nuclear translocators (ARNT1 and ARNT2) that potentially mediate toxic responses to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and related compounds. We have also shown that while both in vitro-expressed chicken AHR1 (ckAHR1) and AHR2 (ckAHR2) exhibit binding affinities to TCDD, only ckAHR1 but not ckAHR2 showed a TCDD-dose-dependent transactivation potency of chicken cytochrome P450 1A5 (ckCYP1A5) in in vitro reporter gene assays. To explore the molecular mechanism of functional difference in the two ckAHRs, the present study investigated the molecular characteristics and function of chicken ARNT (ckARNT) that is a potential dimerization partner for the activation of ckAHR. The full-length ckARNT1 and ckARNT2 cDNAs were isolated and their alternative splice variants were also identified. The ckARNT1 transcript was ubiquitously expressed in various tissues, but ckARNT2 showed restricted expressions in brain, kidney and eye, indicating a similar expression pattern to mammalian ARNTs. The expressions of tagged-ckARNT1 and -ckARNT2 were confirmed in a chicken hepatoma LMH cells by western blot analyses, and their interactions with each ckAHR and a specific recognition DNA element, xenobiotic response element (XRE), were examined by gel shift assays. The result showed that ckARNT1 and ckARNT2 dimerize with each ckAHR isoform and bind with the XRE in a TCDD-dependent manner. Hence, we conclude that functional loss on the dimerization with ckARNTs or the XRE binding is not the major cause of the deficient TCDD-dependency of ckAHR2 for the transactivation. Furthermore, in vitro reporter gene assays showed that transfected ckARNT1 failed to modulate the transcriptional induction of ckAHR-mediated ckCYP1A5 gene by TCDD in COS-7 and LMH cells, whereas ckARNT2 could potentiate the TCDD-dependent response in COS-7 but not in LMH cells. This suggests that ckARNT2 has a distinct role from ckARNT1 in AHR signaling pathway and in a cell-specific mode of action.