A deep learning-based prediction model for prognosis of cervical spine injury: a Japanese multicenter survey

Sadayuki Ito; Hiroaki Nakashima; Naoki Segi; Noriaki Yokogawa; Takeshi Sasagawa; Toru Funayama; Fumihiko Eto; Akihiro Yamaji; Kota Watanabe; Satoshi Nori; Kazuki Takeda; Takeo Furuya; Atsushi Yunde; Hideaki Nakajima; Tomohiro Yamada; Tomohiko Hasegawa; Yoshinori Terashima; Ryosuke Hirota; Hidenori Suzuki; Yasuaki Imajo; Shota Ikegami; Masashi Uehara; Hitoshi Tonomura; Munehiro Sakata; Ko Hashimoto; Yoshito Onoda; Kenichi Kawaguchi; Yohei Haruta; Nobuyuki Suzuki; Kenji Kato; Hiroshi Uei; Hirokatsu Sawada; Kazuo Nakanishi; Kosuke Misaki; Hidetomi Terai; Koji Tamai; Akiyoshi Kuroda; Gen Inoue; Kenichiro Kakutani; Yuji Kakiuchi; Katsuhito Kiyasu; Hiroyuki Tominaga; Hiroto Tokumoto; Yoichi Iizuka; Eiji Takasawa; Koji Akeda; Norihiko Takegami; Haruki Funao; Yasushi Oshima; Takashi Kaito; Daisuke Sakai; Toshitaka Yoshii; Tetsuro Ohba; Bungo Otsuki; Shoji Seki; Masashi Miyazaki; Masayuki Ishihara; Masahiro Oda; Seiji Okada; Shiro Imagama; Satoshi Kato

doi:10.1007/s00586-025-08708-0

A deep learning-based prediction model for prognosis of cervical spine injury: a Japanese multicenter survey

Sadayuki Ito, Hiroaki Nakashima^*, Naoki Segi, Noriaki Yokogawa, Takeshi Sasagawa, Toru Funayama, Fumihiko Eto, Akihiro Yamaji, Kota Watanabe, Satoshi Nori, Kazuki Takeda, Takeo Furuya, Atsushi Yunde, Hideaki Nakajima, Tomohiro Yamada, Tomohiko Hasegawa, Yoshinori Terashima, Ryosuke Hirota, Hidenori Suzuki, Yasuaki ImajoShota Ikegami, Masashi Uehara, Hitoshi Tonomura, Munehiro Sakata, Ko Hashimoto, Yoshito Onoda, Kenichi Kawaguchi, Yohei Haruta, Nobuyuki Suzuki, Kenji Kato, Hiroshi Uei, Hirokatsu Sawada, Kazuo Nakanishi, Kosuke Misaki, Hidetomi Terai, Koji Tamai, Akiyoshi Kuroda, Gen Inoue, Kenichiro Kakutani, Yuji Kakiuchi, Katsuhito Kiyasu, Hiroyuki Tominaga, Hiroto Tokumoto, Yoichi Iizuka, Eiji Takasawa, Koji Akeda, Norihiko Takegami, Haruki Funao, Yasushi Oshima, Takashi Kaito, Daisuke Sakai, Toshitaka Yoshii, Tetsuro Ohba, Bungo Otsuki, Shoji Seki, Masashi Miyazaki, Masayuki Ishihara, Masahiro Oda, Seiji Okada, Shiro Imagama, Satoshi Kato

^*Corresponding author for this work

Orthopaedic Surgery

Research output: Contribution to journal › Article › peer-review

Abstract

Purpose: Cervical spine injuries in the elderly (defined as individuals aged 65 years and older) are increasing, often resulting from falls and minor trauma. Prognosis varies widely, influenced by multiple factors. This study aimed to develop a deep-learning–based predictive model for post-injury outcomes. Methods: This study analyzed a nationwide dataset from the Japan Association of Spine Surgeons with Ambition, comprising 1512 elderly patients (aged 65 years and older) with cervical spine injuries from 2010 to 2020. Deep learning predictive models were constructed for residence, mobility, and the American Spinal Injury Association Impairment Scale (AIS). The model’s performance was compared with that of a traditional statistical analysis. Results: The deep-learning model predicted the residence and AIS outcomes with varying accuracies. The highest accuracy was observed in predicting residence one year post-injury. The model also identified that the AIS score at discharge was significantly predicted by upper extremity trauma, mobility, and elbow extension strength. The deep learning model highlighted factors, such as upper extremity trauma, that were not considered significant in the traditional statistical analysis. Conclusion: Our deep learning-based model offers a novel method for predicting outcomes following cervical spine injuries in the elderly population. The model is highly accurate and provides additional insights into potential prognostic factors. Such models can improve patient care and individualize future interventions.

Original language	English
Journal	European spine journal : official publication of the European Spine Society, the European Spinal Deformity Society, and the European Section of the Cervical Spine Research Society
DOIs	https://doi.org/10.1007/s00586-025-08708-0
State	Accepted/In press - 2025

Keywords

Cervical spine injury
Deep learning
Prediction
Prognosis

ASJC Scopus subject areas

Surgery
Orthopedics and Sports Medicine

Access to Document

10.1007/s00586-025-08708-0

Cite this

Ito, S., Nakashima, H., Segi, N., Yokogawa, N., Sasagawa, T., Funayama, T., Eto, F., Yamaji, A., Watanabe, K., Nori, S., Takeda, K., Furuya, T., Yunde, A., Nakajima, H., Yamada, T., Hasegawa, T., Terashima, Y., Hirota, R., Suzuki, H., ... Kato, S. (Accepted/In press). A deep learning-based prediction model for prognosis of cervical spine injury: a Japanese multicenter survey. European spine journal : official publication of the European Spine Society, the European Spinal Deformity Society, and the European Section of the Cervical Spine Research Society. https://doi.org/10.1007/s00586-025-08708-0

@article{fc8c465812324dfb8d542d42eeb85115,

title = "A deep learning-based prediction model for prognosis of cervical spine injury: a Japanese multicenter survey",

abstract = "Purpose: Cervical spine injuries in the elderly (defined as individuals aged 65 years and older) are increasing, often resulting from falls and minor trauma. Prognosis varies widely, influenced by multiple factors. This study aimed to develop a deep-learning–based predictive model for post-injury outcomes. Methods: This study analyzed a nationwide dataset from the Japan Association of Spine Surgeons with Ambition, comprising 1512 elderly patients (aged 65 years and older) with cervical spine injuries from 2010 to 2020. Deep learning predictive models were constructed for residence, mobility, and the American Spinal Injury Association Impairment Scale (AIS). The model{\textquoteright}s performance was compared with that of a traditional statistical analysis. Results: The deep-learning model predicted the residence and AIS outcomes with varying accuracies. The highest accuracy was observed in predicting residence one year post-injury. The model also identified that the AIS score at discharge was significantly predicted by upper extremity trauma, mobility, and elbow extension strength. The deep learning model highlighted factors, such as upper extremity trauma, that were not considered significant in the traditional statistical analysis. Conclusion: Our deep learning-based model offers a novel method for predicting outcomes following cervical spine injuries in the elderly population. The model is highly accurate and provides additional insights into potential prognostic factors. Such models can improve patient care and individualize future interventions.",

keywords = "Cervical spine injury, Deep learning, Prediction, Prognosis",

author = "Sadayuki Ito and Hiroaki Nakashima and Naoki Segi and Noriaki Yokogawa and Takeshi Sasagawa and Toru Funayama and Fumihiko Eto and Akihiro Yamaji and Kota Watanabe and Satoshi Nori and Kazuki Takeda and Takeo Furuya and Atsushi Yunde and Hideaki Nakajima and Tomohiro Yamada and Tomohiko Hasegawa and Yoshinori Terashima and Ryosuke Hirota and Hidenori Suzuki and Yasuaki Imajo and Shota Ikegami and Masashi Uehara and Hitoshi Tonomura and Munehiro Sakata and Ko Hashimoto and Yoshito Onoda and Kenichi Kawaguchi and Yohei Haruta and Nobuyuki Suzuki and Kenji Kato and Hiroshi Uei and Hirokatsu Sawada and Kazuo Nakanishi and Kosuke Misaki and Hidetomi Terai and Koji Tamai and Akiyoshi Kuroda and Gen Inoue and Kenichiro Kakutani and Yuji Kakiuchi and Katsuhito Kiyasu and Hiroyuki Tominaga and Hiroto Tokumoto and Yoichi Iizuka and Eiji Takasawa and Koji Akeda and Norihiko Takegami and Haruki Funao and Yasushi Oshima and Takashi Kaito and Daisuke Sakai and Toshitaka Yoshii and Tetsuro Ohba and Bungo Otsuki and Shoji Seki and Masashi Miyazaki and Masayuki Ishihara and Masahiro Oda and Seiji Okada and Shiro Imagama and Satoshi Kato",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2025.",

year = "2025",

doi = "10.1007/s00586-025-08708-0",

language = "英語",

journal = "European spine journal : official publication of the European Spine Society, the European Spinal Deformity Society, and the European Section of the Cervical Spine Research Society",

issn = "0940-6719",

publisher = "Springer Verlag",

}

Ito, S, Nakashima, H, Segi, N, Yokogawa, N, Sasagawa, T, Funayama, T, Eto, F, Yamaji, A, Watanabe, K, Nori, S, Takeda, K, Furuya, T, Yunde, A, Nakajima, H, Yamada, T, Hasegawa, T, Terashima, Y, Hirota, R, Suzuki, H, Imajo, Y, Ikegami, S, Uehara, M, Tonomura, H, Sakata, M, Hashimoto, K, Onoda, Y, Kawaguchi, K, Haruta, Y, Suzuki, N, Kato, K, Uei, H, Sawada, H, Nakanishi, K, Misaki, K, Terai, H, Tamai, K, Kuroda, A, Inoue, G, Kakutani, K, Kakiuchi, Y, Kiyasu, K, Tominaga, H, Tokumoto, H, Iizuka, Y, Takasawa, E, Akeda, K, Takegami, N, Funao, H, Oshima, Y, Kaito, T, Sakai, D, Yoshii, T, Ohba, T, Otsuki, B, Seki, S, Miyazaki, M, Ishihara, M, Oda, M, Okada, S, Imagama, S & Kato, S 2025, 'A deep learning-based prediction model for prognosis of cervical spine injury: a Japanese multicenter survey', European spine journal : official publication of the European Spine Society, the European Spinal Deformity Society, and the European Section of the Cervical Spine Research Society. https://doi.org/10.1007/s00586-025-08708-0

A deep learning-based prediction model for prognosis of cervical spine injury: a Japanese multicenter survey. / Ito, Sadayuki; Nakashima, Hiroaki; Segi, Naoki et al.
In: European spine journal : official publication of the European Spine Society, the European Spinal Deformity Society, and the European Section of the Cervical Spine Research Society, 2025.

Research output: Contribution to journal › Article › peer-review

TY - JOUR

T1 - A deep learning-based prediction model for prognosis of cervical spine injury

T2 - a Japanese multicenter survey

AU - Ito, Sadayuki

AU - Nakashima, Hiroaki

AU - Segi, Naoki

AU - Yokogawa, Noriaki

AU - Sasagawa, Takeshi

AU - Funayama, Toru

AU - Eto, Fumihiko

AU - Yamaji, Akihiro

AU - Watanabe, Kota

AU - Nori, Satoshi

AU - Takeda, Kazuki

AU - Furuya, Takeo

AU - Yunde, Atsushi

AU - Nakajima, Hideaki

AU - Yamada, Tomohiro

AU - Hasegawa, Tomohiko

AU - Terashima, Yoshinori

AU - Hirota, Ryosuke

AU - Suzuki, Hidenori

AU - Imajo, Yasuaki

AU - Ikegami, Shota

AU - Uehara, Masashi

AU - Tonomura, Hitoshi

AU - Sakata, Munehiro

AU - Hashimoto, Ko

AU - Onoda, Yoshito

AU - Kawaguchi, Kenichi

AU - Haruta, Yohei

AU - Suzuki, Nobuyuki

AU - Kato, Kenji

AU - Uei, Hiroshi

AU - Sawada, Hirokatsu

AU - Nakanishi, Kazuo

AU - Misaki, Kosuke

AU - Terai, Hidetomi

AU - Tamai, Koji

AU - Kuroda, Akiyoshi

AU - Inoue, Gen

AU - Kakutani, Kenichiro

AU - Kakiuchi, Yuji

AU - Kiyasu, Katsuhito

AU - Tominaga, Hiroyuki

AU - Tokumoto, Hiroto

AU - Iizuka, Yoichi

AU - Takasawa, Eiji

AU - Akeda, Koji

AU - Takegami, Norihiko

AU - Funao, Haruki

AU - Oshima, Yasushi

AU - Kaito, Takashi

AU - Sakai, Daisuke

AU - Yoshii, Toshitaka

AU - Ohba, Tetsuro

AU - Otsuki, Bungo

AU - Seki, Shoji

AU - Miyazaki, Masashi

AU - Ishihara, Masayuki

AU - Oda, Masahiro

AU - Okada, Seiji

AU - Imagama, Shiro

AU - Kato, Satoshi

N1 - Publisher Copyright: © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2025.

PY - 2025

Y1 - 2025

N2 - Purpose: Cervical spine injuries in the elderly (defined as individuals aged 65 years and older) are increasing, often resulting from falls and minor trauma. Prognosis varies widely, influenced by multiple factors. This study aimed to develop a deep-learning–based predictive model for post-injury outcomes. Methods: This study analyzed a nationwide dataset from the Japan Association of Spine Surgeons with Ambition, comprising 1512 elderly patients (aged 65 years and older) with cervical spine injuries from 2010 to 2020. Deep learning predictive models were constructed for residence, mobility, and the American Spinal Injury Association Impairment Scale (AIS). The model’s performance was compared with that of a traditional statistical analysis. Results: The deep-learning model predicted the residence and AIS outcomes with varying accuracies. The highest accuracy was observed in predicting residence one year post-injury. The model also identified that the AIS score at discharge was significantly predicted by upper extremity trauma, mobility, and elbow extension strength. The deep learning model highlighted factors, such as upper extremity trauma, that were not considered significant in the traditional statistical analysis. Conclusion: Our deep learning-based model offers a novel method for predicting outcomes following cervical spine injuries in the elderly population. The model is highly accurate and provides additional insights into potential prognostic factors. Such models can improve patient care and individualize future interventions.

AB - Purpose: Cervical spine injuries in the elderly (defined as individuals aged 65 years and older) are increasing, often resulting from falls and minor trauma. Prognosis varies widely, influenced by multiple factors. This study aimed to develop a deep-learning–based predictive model for post-injury outcomes. Methods: This study analyzed a nationwide dataset from the Japan Association of Spine Surgeons with Ambition, comprising 1512 elderly patients (aged 65 years and older) with cervical spine injuries from 2010 to 2020. Deep learning predictive models were constructed for residence, mobility, and the American Spinal Injury Association Impairment Scale (AIS). The model’s performance was compared with that of a traditional statistical analysis. Results: The deep-learning model predicted the residence and AIS outcomes with varying accuracies. The highest accuracy was observed in predicting residence one year post-injury. The model also identified that the AIS score at discharge was significantly predicted by upper extremity trauma, mobility, and elbow extension strength. The deep learning model highlighted factors, such as upper extremity trauma, that were not considered significant in the traditional statistical analysis. Conclusion: Our deep learning-based model offers a novel method for predicting outcomes following cervical spine injuries in the elderly population. The model is highly accurate and provides additional insights into potential prognostic factors. Such models can improve patient care and individualize future interventions.

KW - Cervical spine injury

KW - Deep learning

KW - Prediction

KW - Prognosis

UR - http://www.scopus.com/inward/record.url?scp=85217569267&partnerID=8YFLogxK

U2 - 10.1007/s00586-025-08708-0

DO - 10.1007/s00586-025-08708-0

M3 - 学術論文

AN - SCOPUS:85217569267

SN - 0940-6719

JO - European spine journal : official publication of the European Spine Society, the European Spinal Deformity Society, and the European Section of the Cervical Spine Research Society

JF - European spine journal : official publication of the European Spine Society, the European Spinal Deformity Society, and the European Section of the Cervical Spine Research Society

ER -

Ito S, Nakashima H, Segi N, Yokogawa N, Sasagawa T, Funayama T et al. A deep learning-based prediction model for prognosis of cervical spine injury: a Japanese multicenter survey. European spine journal : official publication of the European Spine Society, the European Spinal Deformity Society, and the European Section of the Cervical Spine Research Society. 2025. doi: 10.1007/s00586-025-08708-0

A deep learning-based prediction model for prognosis of cervical spine injury: a Japanese multicenter survey

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this