Tomonori Kimura1,2, Ryohei Yamamoto3, Mitsuaki Yoshino4, Ryuichi Sakate4, Enyu Imai5, Shoichi Maruyama6, Hitoshi Yokoyama7, Hitoshi Sugiyama8, Kosaku Nitta9, Tatsuo Tsukamoto10, Shunya Uchida11, Asami Takeda12, Toshinobu Sato13, Takashi Wada14, Hiroki Hayashi15, Yasuhiro Akai16, Megumu Fukunaga17, Kazuhiko Tsuruya18, Kosuke Masutani19, Tsuneo Konta20, Tatsuya Shoji21, Takeyuki Hiramatsu22, Shunsuke Goto23, Hirofumi Tamai24, Saori Nishio25, Kojiro Nagai26, Kunihiro Yamagata27, Hideo Yasuda28, Shizunori Ichida29, Tomohiko Naruse30, Tomoya Nishino31, Hiroshi Sobajima32, Toshiyuki Akahori33, Takafumi Ito34, Yoshio Terada35, Ritsuko Katafuchi36, Shouichi Fujimoto37, Hirokazu Okada38, Tetsushi Mimura39, Satoshi Suzuki40, Yosuke Saka41, Tadashi Sofue42, Kiyoki Kitagawa43, Yoshiro Fujita44, Makoto Mizutani45, Naoki Kashihara46, Hiroshi Sato47, Ichiei Narita48, Yoshitaka Isaka49. 1. Reverse Translational Research Project, Center for Rare Disease Research, National Institutes of Biomedical Innovation, Health and Nutrition (NIBIOHN), Ibaraki, Osaka, Japan. t-kimura@nibiohn.go.jp. 2. Laboratory of Rare Disease Resource Library, Center for Rare Disease Research, National Institutes of Biomedical Innovation, Health and Nutrition (NIBIOHN), Ibaraki, Osaka, Japan. t-kimura@nibiohn.go.jp. 3. Health and Counseling Center, Osaka University, Suita, Osaka, Japan. 4. Reverse Translational Research Project, Center for Rare Disease Research, National Institutes of Biomedical Innovation, Health and Nutrition (NIBIOHN), Ibaraki, Osaka, Japan. 5. Nakayamadera Imai Clinic, Takarazuka, Hyogo, Japan. 6. Department of Nephrology, Nagoya University Graduate School of Medicine, Nagoya, Aichi, Japan. 7. Department of Nephrology, Kanazawa Medical University School of Medicine, Kanazawa, Japan. 8. Department of Nephrology, Rheumatology, Endocrinology and Metabolism, Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, Okayama, Japan. 9. Department of Nephrology, Tokyo Women's Medical University, Tokyo, Japan. 10. Department of Nephrology and Dialysis, Kitano Hospital, Tazuke Kofukai Medical Research Institute, Osaka, Japan. 11. Department of Internal Medicine, Teikyo University School of Medicine, Tokyo, Japan. 12. Kidney Disease Center, Japanese Red Cross Nagoya Daini Hospital, Nagoya, Aichi, Japan. 13. Department of Nephrology, JCHO Sendai Hospital, Sendai, Miyagi, Japan. 14. Department of Nephrology and Laboratory Medicine, Kanazawa University, Kanazawa, Japan. 15. Department of Nephrology, Fujita Health University School of Medicine, Toyoake, Aichi, Japan. 16. First Department of Internal Medicine, Nara Medical University, Nara, Japan. 17. Division of Nephrology, Department of Internal Medicine, Toyonaka Municipal Hospital, Toyonaka, Osaka, Japan. 18. Department of Nephrology, Nara Medical University, Kashihara, Nara, Japan. 19. Division of Nephrology and Rheumatology, Department of Internal Medicine, Faculty of Medicine, Fukuoka University, Fukuoka, Japan. 20. Department of Cardiology, Pulmonology, and Nephrology, Yamagata University School of Medicine, Yamagata, Japan. 21. Department of Kidney Disease and Hypertension, Osaka General Medical Center, Osaka, Japan. 22. Department of Nephrology, Konan Kosei Hospital, Konan, Aichi, Japan. 23. Division of Nephrology and Kidney Center, Kobe University Graduate School of Medicine, Kobe, Hyogo, Japan. 24. Department of Nephrology, Anjo Kosei hospital, Anjo, Aichi, Japan. 25. Division of Rheumatology, Endocrinology and Nephrology, Hokkaido University Graduate School of Medicine, Sapporo, Japan. 26. Department of Nephrology, Institute of Biomedical Sciences, Tokushima University Graduate School, Tokushima, Japan. 27. Department of Nephrology, Faculty of Medicine, University of Tsukuba, Tsukuba, Ibaragi, Japan. 28. Internal Medicine 1, Hamamatsu University School of Medicine, Hamamatsu, Shizuoka, Japan. 29. Department of Nephrology, Japanese Red Cross Nagoya Daiichi Hospital, Nagoya, Aichi, Japan. 30. Department of Nephrology, Kasugai Municipal Hospital, Kasugai, Aichi, Japan. 31. Department of Nephrology, Nagasaki University Hospital, Nagasaki, Japan. 32. Department of Diabetology and Nephrology, Ogaki Municipal Hospital, Ogaki, Ogagki, Japan. 33. Department of Nephrology, Chutoen General Medical Center, Kakegawa, Shizuoka, Japan. 34. Division of Nephrology, Shimane University Hospital, Izumo, Shimane, Japan. 35. Department of Endocrinology, Metabolism and Nephrology, Kochi Medical School, Kochi University, Kochi, Japan. 36. Kideny Unit, National Hospital Organization, Fukuoka-Higashi Medical Center, Koga, Fukuoka, Japan. 37. Department of Hemovascular Medicine and Artificial Organs, Faculty of Medicine, University of Miyazaki, Miyazaki, Japan. 38. Department of Nephrology, Saitama Medical University, Iruma, Saitama, Japan. 39. Department of Nephrology, Gifu Prefectural Tajimi Hospital, Tajimi, Gifu, Japan. 40. Department of Nephrology, Kainan Hospital, Yatomi, Aichi, Japan. 41. Department of Nephrology, Yokkaichi Municipal Hospital, Yokkaichi, Mie, Japan. 42. Department of Cardiorenal and Cerebrovascular Medicine, Kagawa University, Takamatsu, Kagawa, Japan. 43. Division of Internal Medicine, National Hospital Organization Kanazawa Medical Center, Kahoku, Kanazawa, Japan. 44. Department of Nephrology, Chubu Rosai Hospital, Nagoya, Aichi, Japan. 45. Department of Nephrology, Handa City Hospital, Handa, Aichi, Japan. 46. Department of Nephrology and Hypertension, Kawasaki Medical School, Kurashiki, Okayama, Japan. 47. Department of Nephrology, Endocrinology, and Vascular Medicine, Tohoku University Graduate School of Medicine, Sendai, Miyagi, Japan. 48. Division of Clinical Nephrology and Rheumatology, Kidney Research Center, School of Medical and Dental Sciences, Niigata University Graduate, Niigata, Japan. 49. Department of Nephrology, Osaka University Graduate School of Medicine, Suita, Osaka, Japan.
Abstract
BACKGROUND: Prognosis of nephrotic syndrome has been evaluated based on pathological diagnosis, whereas its clinical course is monitored using objective items and the treatment strategy is largely the same. We examined whether the entire natural history of nephrotic syndrome could be evaluated using objective common clinical items. METHODS: Machine learning clustering was performed on 205 cases from the Japan Nephrotic Syndrome Cohort Study, whose clinical parameters, serum creatinine, serum albumin, dipstick hematuria, and proteinuria were traceable after kidney biopsy at 5 measured points up to 2 years. The clinical patterns of time-series data were learned using long short-term memory (LSTM)-encoder-decoder architecture, an unsupervised machine learning classifier. Clinical clusters were defined as Gaussian mixture distributions in a two-dimensional scatter plot based on the highest log-likelihood. RESULTS: Time-series data of nephrotic syndrome were classified into four clusters. Patients in the fourth cluster showed the increase in serum creatinine in the later part of the follow-up period. Patients in both the third and fourth clusters were initially high in both hematuria and proteinuria, whereas a lack of decline in the urinary protein level preceded the worsening of kidney function in fourth cluster. The original diseases of fourth cluster included all the disease studied in this cohort. CONCLUSIONS: Four kinds of clinical courses were identified in nephrotic syndrome. This classified clinical course may help objectively grasp the actual condition or treatment resistance of individual patients with nephrotic syndrome.
BACKGROUND: Prognosis of nephrotic syndrome has been evaluated based on pathological diagnosis, whereas its clinical course is monitored using objective items and the treatment strategy is largely the same. We examined whether the entire natural history of nephrotic syndrome could be evaluated using objective common clinical items. METHODS: Machine learning clustering was performed on 205 cases from the Japan Nephrotic Syndrome Cohort Study, whose clinical parameters, serum creatinine, serum albumin, dipstick hematuria, and proteinuria were traceable after kidney biopsy at 5 measured points up to 2 years. The clinical patterns of time-series data were learned using long short-term memory (LSTM)-encoder-decoder architecture, an unsupervised machine learning classifier. Clinical clusters were defined as Gaussian mixture distributions in a two-dimensional scatter plot based on the highest log-likelihood. RESULTS: Time-series data of nephrotic syndrome were classified into four clusters. Patients in the fourth cluster showed the increase in serum creatinine in the later part of the follow-up period. Patients in both the third and fourth clusters were initially high in both hematuria and proteinuria, whereas a lack of decline in the urinary protein level preceded the worsening of kidney function in fourth cluster. The original diseases of fourth cluster included all the disease studied in this cohort. CONCLUSIONS: Four kinds of clinical courses were identified in nephrotic syndrome. This classified clinical course may help objectively grasp the actual condition or treatment resistance of individual patients with nephrotic syndrome.