[论文解读] Enhancing Psychologists' Understanding through Explainable Deep Learning Framework for ADHD Diagnosis

Abdul Rehman, Ilona Heldal+1 more|arXiv (Cornell University)|Jan 28, 2026

Attention Deficit Hyperactivity Disorder被引用 0

一句话总结

本文提出 HyExDNN-RNN，一种用于 ADHD 检测和多类诊断的可解释深度学习框架，提供 SHAP 与 PFI 解释，在 ADHD200 数据上实现二分类 99% 的准确率和多分类 94.2% 的准确率。

ABSTRACT

Attention Deficit Hyperactivity Disorder (ADHD) is a neurodevelopmental disorder that is challenging to diagnose and requires advanced approaches for reliable and transparent identification and classification. It is characterized by a pattern of inattention, hyperactivity and impulsivity that is more severe and more frequent than in individuals with a comparable level of development. In this paper, an explainable framework based on a fine-tuned hybrid Deep Neural Network (DNN) and Recurrent Neural Network (RNN) called HyExDNN-RNN model is proposed for ADHD detection, multi-class categorization, and decision interpretation. This framework not only detects ADHD, but also provides interpretable insights into the diagnostic process so that psychologists can better understand and trust the results of the diagnosis. We use the Pearson correlation coefficient for optimal feature selection and machine and deep learning models for experimental analysis and comparison. We use a standardized technique for feature reduction, model selection and interpretation to accurately determine the diagnosis rate and ensure the interpretability of the proposed framework. Our framework provided excellent results on binary classification, with HyExDNN-RNN achieving an F1 score of 99% and 94.2% on multi-class categorization. XAI approaches, in particular SHapley Additive exPlanations (SHAP) and Permutation Feature Importance (PFI), provided important insights into the importance of features and the decision logic of models. By combining AI with human expertise, we aim to bridge the gap between advanced computational techniques and practical psychological applications. These results demonstrate the potential of our framework to assist in ADHD diagnosis and interpretation.

研究动机与目标

• 推动更好、可靠的 ADHD 诊断，面向临床心理学家实现透明可解释。
• 开发一个可解释的混合深度学习模型（HyExDNN-RNN），支持二分类 ADHD 检测和多分类分类。
• 结合特征降维与 XAI 方法（SHAP、PFI），提供可解释的诊断洞见。
• 在 ADHD200 数据集上评估模型性能，并评估对临床使用的可解释性。

提出的方法

• 对 HyExDNN-RNN 进行微调，以实现 ADHD 检测与解释。
• 使用 Pearson 相关系数进行特征选择以降维。
• 在降维特征上训练并比较多种模型（RF、DT、EXGB、DNN、LSTM、LSTM-GRU、LSTM-RNN）与 HyExDNN-RNN。
• 应用 SHAP 获得局部特征解释，使用 PFI 获取全局特征重要性以解释决策。
• 采用标准预处理：缺失值填充、标签编码、归一化。
• 使用二元与多类分类指标进行评估（准确率、精确度、召回率、F1-score、ROC-AUC）。

Figure 1 : Proposed XAI-based Machine and Deep Learning Framework for ADHD Diagnosis and Explainability

实验结果

研究问题

• RQ1 HyExDNN-RNN 框架能否在 ADHD 数据集上实现二分类检测和多分类诊断的高准确性？
• RQ2 SHAP 与 PFI 的解释如何揭示特征重要性与模型在 ADHD 诊断中的决策逻辑？
• RQ3 基于 Pearson 的特征选择对模型性能和可解释性的影响如何？
• RQ4 不同模型在 ADHD 检测的准确性与鲁棒性方面与 HyExDNN-RNN 相比如何？

主要发现

• HyExDNN-RNN 在二分类中达到 99% 的准确率和 0.99 的 ROC-AUC。
• HyExDNN-RNN 在多分类中达到 94.2% 的准确率、精确度、召回率与 F1-score。
• SHAP 识别注意力不集中、过度活跃/冲动 ADHD 指标为对 ADHD 预测高度影响的特征。
• PFI 提供全球特征重要性，揭示跨预测的整体模型行为。
• EXGB 与 LSTM-RNN 也表现良好，但 HyExDNN-RNN 显示出最强的综合表现与可解释性。
• 该框架展示了如何让 AI 搭配人类专业知识来支持 ADHD 诊断与解释。

Figure 2 : Distribution of Dataset Classes. Typically Developing Children (TDC) Class 0, ADHD-combined (Class 1), ADHD-Hyperactive/Impulsive (Class 2), ADHD-Inattentive (Class 3)