Background It is expected that artificial intelligence (AI) will be used extensively in the medical field in the future. Objective The purpose of this study is to investigate the awareness of AI among Korean doctors and to assess physicians’ attitudes toward the medical application of AI. Methods We conducted an online survey composed of 11 closed-ended questions using Google Forms. The survey consisted of questions regarding the recognition of and attitudes toward AI, the development direction of AI in medicine, and the possible risks of using AI in the medical field. Results A total of 669 participants completed the survey. Only 40 (5.9%) answered that they had good familiarity with AI. However, most participants considered AI useful in the medical field (558/669, 83.4% agreement). The advantage of using AI was seen as the ability to analyze vast amounts of high-quality, clinically relevant data in real time. Respondents agreed that the area of medicine in which AI would be most useful is disease diagnosis (558/669, 83.4% agreement). One possible problem cited by the participants was that AI would not be able to assist in unexpected situations owing to inadequate information (196/669, 29.3%). Less than half of the participants(294/669, 43.9%) agreed that AI is diagnostically superior to human doctors. Only 237 (35.4%) answered that they agreed that AI could replace them in their jobs. Conclusions This study suggests that Korean doctors and medical students have favorable attitudes toward AI in the medical field. The majority of physicians surveyed believed that AI will not replace their roles in the future.

PurposeThis study was performed to investigate the incidence and configuration of the bifid mandibular canal in a Korean population by using cone-beam computed tomography (CBCT) imaging.Materials and MethodsCBCT images of 1933 patients (884 male and 1049 female) were evaluated using PSR-9000N and Alphard-Vega 3030 Dental CT units (Asahi Roentgen Ind. Co., Ltd, Kyoto, Japan). Image analysis was performed by using OnDemand3D software (CyberMed Inc., Seoul, Korea). The bifid mandibular canal was identified and classified into four types, namely, the forward canal, buccolingual canal, dental canal, and retromolar canal. Statistical analysis was performed by using the chi-squared test and one-way analysis of variance (ANOVA).ResultsBifid mandibular canals were observed in 198 (10.2%) of 1933 patients. The most frequently observed type of bifid mandibular canal was the retromolar canal (n=104, rate: 52.5%) without any significant difference among the incidence of each age and gender. The mean diameter of the accessory canal was 1.27 mm (range: 0.27-3.29 mm) without any significant difference among the mean diameter of each type of the bifid mandibular canal. The mean length of the bifid mandibular canals was 14.97mm(range: 2.17-38.8 mm) with only a significant difference between the dental canal and the other types.ConclusionThe bifid mandibular canal is not uncommon in Koreans and has a prevalence of 10.2% as indicated in the present study. It is suggested that a CBCT examination be recommended for detecting a bifid canal.

No abstract

Primary intraosseous squamous cell carcinoma (PIOSCC) is a rare carcinoma, which arises within the jaws without connection to the oral mucosa and presumably develops from a remnant of odontogenic epithelium. We present a case of solid type PIOSCC in a 52-year-old male patient complaining of dull pain on his left lower molar. In this case, early stage PIOSCC mimicking a periapical lesion might lead to a one-year delay in treatment due to the misdiagnosis of osteomyelitis after extraction of the third molar. The clinical, radiological, and histologic features are described. In this case, there was initial radiographic evidence for PIOSCC mimicking a periapical lesion. Incautious radiographic interpretation and treatment procedures had delayed the correct diagnosis and resulted in extensive bony destruction during the patient's disease progression.

Background Fabry disease is a rare X-linked genetic lysosomal disorder caused by mutations in the GLA gene encoding alpha-galactosidase A. Despite some data showing that profibrotic and proinflammatory cytokines and oxidative stress could be involved in Fabry disease-related renal injury, the pathogenic link between metabolic derangement within cells and renal injury remains unclear. Methods Renal fibrosis was triggered by unilateral ureteral obstruction (UUO) in mice with Fabry disease to investigate the pathogenic mechanism leading to fibrosis in diseased kidneys. Results Compared to kidneys of wild-type mice, lamellar inclusion bodies were recognized in proximal tubules of mice with Fabry disease. Sirius red and trichrome staining revealed significantly increased fibrosis in all UUO kidneys, though it was more prominent in obstructed Fabry kidneys. Renal messenger RNA levels of inflammatory cytokines and profibrotic factors were increased in all UUO kidneys compared to sham-operated kidneys but were not significantly different between UUO control and UUO Fabry mice. Protein levels of Nox2, Nox4, NQO1, catalase, SOD1, SOD2, and Nrf2 were not significantly different between UUO control and UUO Fabry kidneys, while the protein contents of LC3-II and LC3-I and expression of Beclin1 were significantly decreased in UUO kidneys of Fabry disease mouse models compared with wild-type mice. Notably, TUNEL-positive cells were elevated in obstructed kidneys of Fabry disease mice compared to wild-type control and UUO mice. Conclusion These findings suggest that impaired autophagy and enhanced apoptosis are probable mechanisms involved in enhanced renal fibrosis under the stimulus of UUO in Fabry disease.

No abstract

BACKGROUND It is expected that artificial intelligence (AI) will be used extensively in the medical field in the future. OBJECTIVE The purpose of this study is to investigate the awareness of AI among Korean doctors and to assess physicians’ attitudes toward the medical application of AI. METHODS We conducted an online survey composed of 11 closed-ended questions using Google Forms. The survey consisted of questions regarding the recognition of and attitudes toward AI, the development direction of AI in medicine, and the possible risks of using AI in the medical field. RESULTS A total of 669 participants completed the survey. Only 40 (5.9%) answered that they had good familiarity with AI. However, most participants considered AI useful in the medical field (558/669, 83.4% agreement). The advantage of using AI was seen as the ability to analyze vast amounts of high-quality, clinically relevant data in real time. Respondents agreed that the area of medicine in which AI would be most useful is disease diagnosis (558/669, 83.4% agreement). One possible problem cited by the participants was that AI would not be able to assist in unexpected situations owing to inadequate information (196/669, 29.3%). Less than half of the participants(294/669, 43.9%) agreed that AI is diagnostically superior to human doctors. Only 237 (35.4%) answered that they agreed that AI could replace them in their jobs. CONCLUSIONS This study suggests that Korean doctors and medical students have favorable attitudes toward AI in the medical field. The majority of physicians surveyed believed that AI will not replace their roles in the future.

No abstract