Seung Ho Choi

Sybil, an open-source deep learning model, predicted future lung cancer risk in Asian individuals with heavy smoking histories of at least 20 pack-years using low-dose CT scans, supporting its role in risk-based screening strategies, but demonstrated poor performance in a never- or light-smoking subgroup (median, 2 pack-years).

In this retrospective study of individuals from the Korean National Lung Cancer Screening Program, presence of interstitial lung abnormalities was associated with higher risks of lung cancer, idiopathic pulmonary fibrosis, and mortality.

The open-source artificial intelligence CXR-Lung-Risk model predicted respiratory disease mortality in an Asian cohort using baseline and follow-up chest radiographs, facilitating a two-layer risk stratification approach through longitudinal analysis.

OBJECTIVES: Alcohol consumption is causally linked to several cancers, and major health organizations classify it as a carcinogen. This study assessed the impact of alcohol consumption on cancer incidence and mortality in Korea in 2015 and 2020, projected trends up to 2030, and compared results based on different criteria.METHODS: The relative risk of cancer associated with alcohol consumption in Korea was determined through a meta-analysis of alcohol-related relative risks for specific cancers, using primary data from the Korean Cohort Study within the Korean Cohort Consortium. The population-attributable fraction (PAF) was calculated using Levin’s formula, incorporating drinking prevalence and the number of cancer cases and deaths, with a 15-year latency period assumed.RESULTS: In Korea, the PAF for alcohol consumption, based on ever/never drinking criteria, was higher than that calculated using other criteria, except for the PAF based on past and current/never drinking criteria. Alcohol consumption contributed to 3.58% of all cancer cases and 3.28% of cancer deaths in 2015. It accounted for 4.58% of new cancer cases in male and 2.08% in female, with a higher contribution to incidence than mortality (4.00 and 2.25% of cancer deaths in male and female, respectively). Projections indicate that alcohol-related cancer PAF will decrease by 17.2% in male but increase by 70.2% in female by 2030.CONCLUSIONS: This study highlights the impact of alcohol consumption on cancer in Korea, emphasizing the need for sex-specific regulations to address sex differences.

OBJECTIVES: Tobacco smoking is a major public health concern worldwide. This study aimed to assess its impact on cancer incidence and mortality by estimating the population attributable fraction (PAF) in the Korean population for 2015 and 2020 and by projecting future trends until 2030.METHODS: The Korean relative risk (RR) was calculated via a meta–analysis of RRs for individual cancers attributed to tobacco smoking, based on primary data analysis from the Korean Cohort Consortium. The PAF was estimated using the Levin formula with past and current prevalence rates and the number of cancer cases and deaths, assuming a 15-year latency period.RESULTS: The proportions of cancer cases and deaths in Korea attributable to tobacco smoking were similar to those calculated using Asian and global RRs for both male and female. In 2015 and 2020, tobacco smoking contributed to 14.32% and 13.17% of cancer cases and 21.70% and 20.69% of cancer deaths in adults, respectively. Among Koreans, smoking was responsible for 25.83% of new cancer cases in male in 2015, 23.49% in male in 2020, 1.46% in female in 2015, and 1.68% in female in 2020. In both years, smoking impacted mortality more strongly than incidence in Korean male and female (incidence in male: 25.83% and 23.49%; mortality in male: 32.09% and 30.41%; incidence in female: 1.46% and 1.68%; and mortality in female: 4.70% and 4.96%, respectively).CONCLUSIONS: Tobacco smoking causes cancers and deaths in Korea, however, it is preventable. Effective control policies that consider trends and vulnerabilities among female are required.

OBJECTIVES: This study aimed to determine the population-attributable fractions (PAFs) of cancers using various calculation methods and to estimate the PAFs of cancer incidence and mortality resulting from deficit in physical activity (DPA) from 2015 to 2030, based on data on prevalence rates.METHODS: The PAF of cancer was estimated using a cohort study-based meta-analysis of relative risk (RR), national prevalence rates of DPA from 2000 to 2015, and national cancer statistics from 2015 to 2030, with a latency of 15 years.RESULTS: In 2015, DPA contributed to 909 cancer cases and 548 deaths, accounting for 0.42% and 0.68% of new cancer cases and deaths, respectively. By 2030, the PAF values are expected to increase to 1.31% of incidence and 1.80% of mortality, with a continual increase from 2015 to 2030. When the low metabolic equivalent of task (MET) criteria were selected, the PAF values decreased for both incidence and mortality. The PAF calculated with <900 MET-min/wk for the sex-specific MET criterion was higher than that calculated with <900 MET-min/wk for both incidence and mortality.CONCLUSIONS: The risk of cancer associated with DPA is expected to rise in both male and female. Future research and strategies should emphasize the promotion of physical activity for cancer prevention, considering its significant implications for public health.

"Just Accepted" papers have undergone full peer review and have been accepted for publication in Radiology: Artificial Intelligence. This article will undergo copyediting, layout, and proof review before it is published in its final version. Please note that during production of the final copyedited article, errors may be discovered which could affect the content. Purpose To assess the prognostic value of a deep learning-based chest radiographic age (CXR-Age) model in a large external test cohort of Asian individuals. Materials and Methods This single-center, retrospective study included chest radiographs from consecutive, asymptomatic Asian individuals aged 50 to 80 who received health check-ups between January 2004 and June 2018. This study performed a dedicated external test of the previously developed CXR-Age model, which predicts an age adjusted based on the risk of all-cause mortality. Adjusted hazard ratios (HRs) of CXR-Age for all-cause, cardiovascular, lung cancer, and respiratory disease mortality were assessed using multivariable Cox or Fine-Gray models, with their added values evaluated by likelihood ratio tests. Results A total of 36,924 individuals (mean chronological age ± SD, 58 ± 7 years; CXR-Age, 60 ± 5 years; 22,352 male) were included. Over a median follow-up of 11.0 years, 1250 individuals (3.4%) died, including 153 cardiovascular (0.4%), 166 lung cancer (0.4%), and 98 respiratory deaths (0.3%). CXR-Age was a significant risk factor for all-cause (adjusted HR at the chronological age of 50 years: 1.03; at 60 years: 1.05; at 70 years: 1.07), cardiovascular (adjusted HR: 1.11), lung cancer (adjusted HR for former smokers: 1.12; for current smokers: 1.05), and respiratory disease mortality (adjusted HR: 1.12) (all P values < 0.05). The likelihood ratio test demonstrated added prognostic value of CXR-Age to clinical factors including chronological age for all outcomes (all P values < 0.001). Conclusion Deep learning-based chest radiographic age was associated with various survival outcomes and had added value to clinical factors in asymptomatic Asian individuals, suggesting its generalizability. ©RSNA, 2024.

Huiyul Park, Eileen L. Yoon, Goh Eun Chung, Eun Kyung Choe, Jung Ho Bae, Seung Ho Choi, Mimi Kim, Woochang Hwang, Hye-Lin Kim, Sun Young Yang, Dae Won Jun. Gut and Liver 2024;18:316-27. https://doi.org/10.5009/gnl230044

A deep learning model to predict incident lung cancer from chest radiographs showed added value to the 2021 U.S. Preventive Services Task Force recommendations for low-dose CT lung cancer screening... Background A deep learning (DL) model to identify lung cancer screening candidates based on their chest radiographs requires external validation with a recent real-world non–U.S. sample. Purpose To...