Construction of Prediction Model for Brain Strain Waveforms in Vehicle Crash Tests using Deep Learning
深層学習による自動車衝突時の脳ひずみ波形予測モデルの構築
- Delivery
- Available on the other site
- Click here to order.
- Publication code
- 20244229
- Paper/Info type
- JSAE Transaction
Vol.55 No.3
- Pages
- 565-570(Total 6 p)
- Date of publication
- May 2024
- Publisher
- JSAE
- Language
- Japanese
Detailed Information
| Category(J) | 論文 Translation |
|---|---|
| Category(E) | Paper |
| Author(J) | 1) 玉井 駿太朗, 2) 宮崎 祐介 |
| Author(E) | 1) Shuntaro Tamai, 2) Yusuke Miyazaki |
| Affiliation(J) | 1) 東京工業大学, 2) 東京工業大学 |
| Abstract(J) | 自動車衝突試験によって得られた頭部角速度波形および有限要素解析によって取得した脳内ひずみ波形を用い，深層学習による自動車衝突時の脳ひずみ波形予測モデルを構築した．予測モデル構築過程では，力学的要因を考慮に入れた説明変数の作成とデータ拡張手法を導入することで，モデルの予測性能を向上させた． Translation |
| Abstract(E) | Brain strain indices calculated using human finite element models have recently been used to evaluate the brain injury risk in automobile accidents. However, the calculation using a finite element model of a human body is computationally expensive and time-consuming, making it impossible to evaluate brain strain indices immediately after a crash test. Therefore, this paper develops a deep learning model to predict the brain strain waveform from the angular velocity waveform of the head that can be measured by a crash dummy. The results of a comparison between the brain strain waveform obtained by finite element analysis and the waveform predicted by the developed deep learning model showed that the CORA evaluation exceeded 0.9, indicating that the model predicted the waveform with a high degree of accuracy. |
