2022FinTechathon深圳国际金融科技大赛（人工智能赛道 ）.zip

# 2022FinTechathon 深圳国际金融科技大赛 ### Fed-heathcare Team Won the 2022 FinTechathon Competition First Prize! 团队：Fed-healthcare (联邦医疗) 面向智能体音感知的联邦学习系统-基于FATE平台  # FederatedHealth_HeartSound The horizontal FL (HFL) and vertical FL (VFL) paradigms for heart sound analysis. #### Index Terms— Computer audition, federated learning, heart sound, information security, model interpretability # Algorithm_Models * Paradigms and workflows of horizontal and vertical federated learning (FL) on multi-institutional heart sound databases. #### **Note:** We first provide the final models and the relevant experimental results for this challenge. Other major program files will be uploaded after the paper of "Heart Sound Abnormality Detection from Multi-institutional Collaboration: Introducing a Federated Ensemble Learning Framework" is accepted. ## Results * Horizontally-Federated Learning vs Data-Centralised Learning Table 1. A SUMMARY OF RESULTS (IN [%]) FOR CLASSIC XGBOOST AND THE HFL MODEL WITH OPTIMAL PARAMETERS. | | Acc | Se | Sp | UF1 | UAR | | ----- | ----- | ---- |---- |---- |---- | | XGBoost(Centralised Data)| 68.4 | 69.1 |67.6 | 68.4 | 68.4 | | Homogeneous-SecureBoost | 67.5 | 62.1 |72.8 | 67.4 | 67.5 | Important parameters settings for the HFL and the XGBoost: tree depth=3, tree number=30, subsample feature rate=1.0, learning rate=0.3.  Fig. 1. Normalised confusion matrix (in [%]) for the XGBoost and Horizontal-SecureBoost models. Table 2. COMPARISON OF THE RESULTS (IN [%]) OF THE CONVENTIONAL XGBOOST AND HETEROGENEOUS SECUREBOOST MODELS ON DATA FOR EACH INSTITUTION. | | XGBoost(Centralised Data)| |Heterogeneous-SecureBoost| | | | Acc | Se | Sp | UAR | Acc | Se | Sp | UAR | | ----- | ----- | ---- |---- |---- |---- |----- | ---- |---- | | Db | 86.7 |85.2 |88.3 |86.8 |82.7 |82.0 |83.5 |82.7 | | Dc | 86.7 |85.7 |87.5 |86.6 |93.3 |85.7 |92.0 |92.9 | | Dd | 93.3 |87.5 |92.0 |93.8 |96.2 |89.6 |96.4 |97.2 | | De | 88.2 |85.6 |86.7 |87.8 |87.6 |82.6 |86.3 |84.3 | | Df | 86.8 |90.9 |81.3 |86.1 |79.5 |75.5 |71.3 |78.4 |  Fig. 2. Normalised confusion matrix (in [%]) for XGBoost and Vertically- SecureBoost models trained at D_e database. ## Awards   Fig. 3. 2022Fintechathon Shenzhen International FinTechathon Prize. https://www.infoq.cn/zones/fintechathon/campus2022/result ## Availability 1. Voice of the Body (VoB) 是第一个计算机听觉医学数据库平台，用于对体音信号进行分析. https://www.vob-bit.org/ 2. Classification of Heart Sound Recordings (PhysioNet/CinC challenge): https://physionet.org/content/challenge-2016/1.0.0. 3. SHAP (SHapley Additive exPlanations) is a game-theoretic method to explain the output of ML models. https://shap.readthedocs.io. 4. FATE (Federated AI Technology Enabler) supports the FL architecture, as well as the secure computation and development of various ML algorithms. https://github.com/FederatedAI/FATE. ## Cite As Wanyong Qiu, Chen Quan, Lixian Zhu, Yongzi Yu, Zhihua Wang, Yu Ma, Mengkai Sun, Yi Chang, Kun Qian*, Bin Hu∗, Yoshiharu Yamamoto and Bjoern W. Schuller, “Heart Sound Abnormality Detection from Multi-institutional Collaboration: Introducing a Federated Ensemble Learning Framework”, IEEE IoTJ, pp. 1-11, Submitted, July 2023.