Deep Learning Versus LASSO-Based Machine Learning for 1-Year Survival After Surgery for Spinal Metastases

Study Design. Multicenter prospective cohort study; secondary analysis. Objective. To evaluate predictors associated with 1-year survival after surgery for spinal metastases by comparing a comprehensive 50-variable deep learning (DL) model with a previously published 5-variable LASSO-based machine learning (ML) model and applying DL-based permutation feature importance as an exploratory analytic lens. Summary of Background Data. Surgical decision-making for spinal metastases requires reliable survival estimates. Traditional scores such as those of Tokuhashi and Tomita and contemporary tools such as SORG and NESMS support prognostication, but performance and calibration may vary across cohorts. A parsimonious 5-variable JASA ML model is clinically practical, whereas DL may help identify prognostic signals embedded in detailed activities of daily living (ADLs), patient-reported outcomes (PROs), and scoring-system components. Methods. We analyzed 401 complete-case patients who underwent surgery for spinal metastases at 35 Japanese institutions (2018-2021). A feed-forward neural network incorporating 50 preoperative variables was evaluated using five repeated random 8:2 train-test splits. Accuracy, AUROC, Brier score, and calibration summaries were reported and descriptively compared with the previously published 5-variable LASSO-based ML model. Results. At 1 year, 269 of 401 patients were alive. The DL model achieved 75.5+/- 3.0% accuracy (95% confidence interval [CI], 71.8%-79.2%), held-out AUROC 0.789 (95% CI, 0.681-0.886), and Brier score 0.214. The ML model achieved 71.8% accuracy (Wilson 95% CI, 67.2%-76.0%) and apparent AUROC 0.762. Because the comparison was descriptive rather than paired, formal statistical superiority was not claimed. DL feature importance highlighted Vitality Index-On and Off Toilet, EQ-5D-5L total score and pain/discomfort, and individual Tokuhashi/Tomita components; the ML-selected Vitality Index-Wake Up item ranked 38th. Conclusions. The 50-variable DL model provided reasonable prediction and generated clinically plausible feature-importance hypotheses, but it did not demonstrate a clearly meaningful performance advantage over the simpler 5-variable ML model. DL may be most useful for research-based feature discovery and refinement of future parsimonious prognostic tools, whereas validated simple models remain more practical for bedside prognostication. Level of Evidence. 2

Authors

Institutions

Publication Details

Journal
Spine
Published
2026-08-26
DOI
https://doi.org/10.1097/brs.0000000000005843
Primary Topic
Management of metastatic bone disease
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
article

Deep Learning Versus LASSO-Based Machine Learning for 1-Year Survival After Surgery for Spinal Metastases

Masashi Miyazaki, Tsutomu Oshigiri, Kazuyuki Watanabe, Jun Ouchida et al.
Spine
Management of metastatic bone disease
article

Deep Learning Versus LASSO-Based Machine Learning for 1-Year Survival After Surgery for Spinal Metastases

Masashi Miyazaki, Tsutomu Oshigiri, Kazuyuki Watanabe, Jun Ouchida, Masahiro Funaba, Yutaro Kanda, Yohei Takahashi, Chizuo Iwai, Toshio Nakamae, Takuya Iimura, Hirokatsu Sawada, Eiki Shirasawa, Sadayuki Ito, Koji Uotani, Yuki Shiratani, Hidenori Suzuki, Masaaki Paku, Hirokazu Inoue, Toru Funayama, Naoki Segi, Daisuke Yamabe, Hiroyuki Tominaga, Norihiko Takegami, Yuta Goto, Ichiro Kawamura, Koji Matsumoto, Kousei Miura, Hiroaki Manabe, Shuji Watanabe, Ko Hashimoto, Shinji Tanishima, Hiroaki Nakashima, Narihito Nagoshi, Koji Akeda, Masayuki Ishihara, Atsushi Kimura, Takaki Shimizu, Satoshi Kato, Kazuo Nakanishi, Takashi Kaito, Akihiko Hiyama, Gen Inoue, Shoji Seki, Kenichiro Kakutani, Kota Watanabe, Shiro Imagama, Hiroshi Moridaira, Bungo Otsuki, Takashi Hirai, Kazu Kobayakawa, Haruki Funao, Hideaki Nakajima, Takeo Furuya, Hidetomi Terai, Akinobu Suzuki
article en

Abstract

Study Design. Multicenter prospective cohort study; secondary analysis. Objective. To evaluate predictors associated with 1-year survival after surgery for spinal metastases by comparing a comprehensive 50-variable deep learning (DL) model with a previously published 5-variable LASSO-based machine learning (ML) model and applying DL-based permutation feature importance as an exploratory analytic lens. Summary of Background Data. Surgical decision-making for spinal metastases requires reliable survival estimates. Traditional scores such as those of Tokuhashi and Tomita and contemporary tools such as SORG and NESMS support prognostication, but performance and calibration may vary across cohorts. A parsimonious 5-variable JASA ML model is clinically practical, whereas DL may help identify prognostic signals embedded in detailed activities of daily living (ADLs), patient-reported outcomes (PROs), and scoring-system components. Methods. We analyzed 401 complete-case patients who underwent surgery for spinal metastases at 35 Japanese institutions (2018-2021). A feed-forward neural network incorporating 50 preoperative variables was evaluated using five repeated random 8:2 train-test splits. Accuracy, AUROC, Brier score, and calibration summaries were reported and descriptively compared with the previously published 5-variable LASSO-based ML model. Results. At 1 year, 269 of 401 patients were alive. The DL model achieved 75.5+/- 3.0% accuracy (95% confidence interval [CI], 71.8%-79.2%), held-out AUROC 0.789 (95% CI, 0.681-0.886), and Brier score 0.214. The ML model achieved 71.8% accuracy (Wilson 95% CI, 67.2%-76.0%) and apparent AUROC 0.762. Because the comparison was descriptive rather than paired, formal statistical superiority was not claimed. DL feature importance highlighted Vitality Index-On and Off Toilet, EQ-5D-5L total score and pain/discomfort, and individual Tokuhashi/Tomita components; the ML-selected Vitality Index-Wake Up item ranked 38th. Conclusions. The 50-variable DL model provided reasonable prediction and generated clinically plausible feature-importance hypotheses, but it did not demonstrate a clearly meaningful performance advantage over the simpler 5-variable ML model. DL may be most useful for research-based feature discovery and refinement of future parsimonious prognostic tools, whereas validated simple models remain more practical for bedside prognostication. Level of Evidence. 2

Spine
Kanazawa University (JP), Nihon University (JP), Kagoshima University (JP), University of Fukui (JP), Hiroshima University (JP), Tokyo Medical and Dental University (JP), Kansai Medical University (JP), Fukushima Medical University (JP), Tokai University (JP), Kyushu University (JP), Sapporo Medical University (JP), University of Tsukuba (JP), Jichi Medical University (JP), Yamaguchi University (JP), Mie University (JP), Iwate Medical University (JP), Oita University (JP), Kawasaki Medical School (JP), Tohoku University (JP), Keio University (JP), Nagoya City University (JP), Chiba University Hospital (JP), Okayama University Hospital (JP), Gifu University Hospital (JP), Keio University Hospital (JP), Kyoto University Hospital (JP), Jichi Medical University Hospital (JP), University of Toyama (JP), Osaka Metropolitan University (JP), Tottori University (JP), Nagoya University (JP), Kitasato University (JP), Kobe University (JP), The University of Tokyo (JP), Dokkyo Medical University (JP), International University of Health and Welfare (JP), Tokushima University (JP), The University of Osaka (JP)
Peace, Justice and strong institutions
Openalex Percentile: Top 8%
Management of metastatic bone disease
DOISourceOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.