- 著者
- Yuki Kataoka Tomohisa Baba Tatsuyoshi Ikenoue Yoshinori Matsuoka Junichi Matsumoto Junji Kumasawa Kentaro Tochitani Hiraku Funakoshi Tomohiro Hosoda Aiko Kugimiya Michinori Shirano Fumiko Hamabe Sachiyo Iwata Yoshiro Kitamura Tsubasa Goto Tomohiro Handa Shoji Kido Shingo Fukuma Noriyuki Tomiyama Toyohiro Hirai Takashi Ogura Japan COVID-19 AI team
- 出版者
- Society for Clinical Epidemiology
- 雑誌
- Annals of Clinical Epidemiology (ISSN:24344338)
- 巻号頁・発行日
- pp.22014, (Released:2022-07-08)
- 被引用文献数
- 2
Background: We aimed to develop and externally validate a novel machine learning model that can classify CT image findings as positive or negative for SARS-CoV-2 reverse transcription polymerase chain reaction (RT-PCR).Methods: We used 2,928 images from a wide variety of case-control type data sources for the development and internal validation of the machine learning model. A total of 633 COVID-19 cases and 2,295 non-COVID-19 cases were included in the study. We randomly divided cases into training and tuning sets at a ratio of 8:2. For external validation, we used 893 images from 740 consecutive patients at 11 acute care hospitals suspected of having COVID-19 at the time of diagnosis. The dataset included 343 COVID-19 patients. The reference standard was RT-PCR.Results: In external validation, the sensitivity and specificity of the model were 0.869 and 0.432, at the low-level cutoff, 0.724 and 0.721, at the high-level cutoff. Area under the receiver operating characteristic was 0.76.Conclusions: Our machine learning model exhibited a high sensitivity in external validation datasets and may assist physicians to rule out COVID-19 diagnosis in a timely manner at emergency departments. Further studies are warranted to improve model specificity.
- 著者
- Shingo Fukuma Tatsuyoshi Ikenoue Yukari Yamada Yoshiyuki Saito Joseph Green Takeo Nakayama Shunichi Fukuhara
- 出版者
- Japan Epidemiological Association
- 雑誌
- Journal of Epidemiology (ISSN:09175040)
- 巻号頁・発行日
- vol.31, no.7, pp.410-416, 2021-07-05 (Released:2021-07-05)
- 参考文献数
- 39
- 被引用文献数
- 3
Background: Breaches of ethics undermine the practice of medicine. In Japan, two major scandals involving clinical research and drug marketing occurred after the publication of clinical trials. To study the effects of those scandals, we evaluated changes in the use of first-generation angiotensin II receptor blockers (ARBs) after publication of relevant clinical trials and also after the subsequent scandals.Methods: We conducted a quasi-experimental design of an interrupted time series analysis (ITSA) on nationwide monthly drug-market data covering 12 years (2005 to 2017) in Japan. The main outcome was the use of first-generation ARBs (valsartan, candesartan, and losartan). The two exposures were the publication of ARB-related clinical-trial results (October 2006) and subsequent ARB-related scandals involving research and marketing (February 2013). A generalized estimating equation model was fitted for ITSA with a log link, Poisson distribution, robust variance estimators, and seasonality adjustment.Results: The publication of clinical trials was associated with 12% increase in the use of first-generation ARBs in Japan, and the subsequent ARB-related scandals was associated with 19% decrease. The decrease in the use of first-generation ARBs after the scandals was greater than the increase in their use after the publication of clinical-trial results. The net effect of the two exposures was a 9% decrease in the use of first-generation ARBs.Conclusions: The scandals were associated with decrease in the use of first-generation ARBs, and that decrease was greater than the increase associated with the publication of “successful” clinical trials, making the net effect not zero but negative.
- 著者
- Shingo Fukuma Tatsuyoshi Ikenoue Yukari Yamada Yoshiyuki Saito Joseph Green Takeo Nakayama Shunichi Fukuhara
- 出版者
- Japan Epidemiological Association
- 雑誌
- Journal of Epidemiology (ISSN:09175040)
- 巻号頁・発行日
- pp.JE20200181, (Released:2020-07-04)
- 参考文献数
- 39
- 被引用文献数
- 3
BackgroundBreaches of ethics undermine the practice of medicine. In Japan, two major scandals involving clinical research and drug marketing occurred after the publication of clinical trials. To study the effects of those scandals, we evaluated changes in the use of first-generation angiotensin II receptor blockers (ARBs) after publication of relevant clinical trials and also after the subsequent scandals.MethodsWe conducted a quasi-experimental design of an interrupted time series analysis (ITSA) on nationwide monthly drug-market data covering 12 years (2005 to 2017) in Japan. The main outcome was the use of first-generation ARBs (valsartan, candesartan, and losartan). The two exposures were the publication of ARB-related clinical-trial results (October 2006) and subsequent ARB-related scandals involving research and marketing (February 2013). A generalized estimating equation model was fitted for ITSA with a log link, Poisson distribution, robust variance estimators, and seasonality adjustment.ResultsThe publication of clinical trials was associated with 12% increase in the use of first-generation ARBs in Japan, and the subsequent ARB-related scandals was associated with 19% decrease. The decrease in the use of first-generation ARBs after the scandals was greater than the increase in their use after the publication of clinical-trial results. The net effect of the two exposures was a 9% decrease in the use of first-generation ARBs.ConclusionsThe scandals were associated with decrease in the use of first-generation ARBs, and that decrease was greater than the increase associated with the publication of “successful” clinical trials, making the net effect not zero but negative.