HAM10000 expert-annotated explanations and pseudonymized reader study data.

Reader study was conducted in 3 phases: no ai (phase 1), ai support (phase 2), xai support (phase 3). The data generated in each phase is available in metadata_phase1.csv, metadata_phase2.csv, and metadata_phase3.csv.There are 113 unique participants in phase 1. Additional participants were added in phase 2, resulting in 116 unique clinicians in phases 2 and 3.<b>Important:</b> The 3rd and 13th image in each group are identical. Be careful when performing table joins as the duplicate image_ids can affect them. In metadata_phase1.csv, the AI predictions for the 13th image in each group are null. Please take that into account when performing analysis. In metadata_phase2 and metadata_phase3, the AI predictions for the repeating images are not omitted.<br><i>participant</i>: Each clinician was assigned a participant Id represented by the participant column.<i>group</i>: Each clinician was randomly assigned to a group. Each group was assigned mutually exclusive sets of images.<i>mask</i>: An internal identifier used for the images. Can be ignored.<i>benign_malignant</i>: ground truth diagnosis.<i>prediction</i>: Diagnosis chosen by the clinician. 1 represents melanoma, 0 represents nevus, 0.5 represents a nevus diagnosis but the clinician chose to excise.<i>confidence</i>: Confidence value entered by the clinician.<i>trust</i>: Trust value entered by the clinician.<i>AI_prediction</i>: Diagnosis predicted by the AI. 1 represents melanoma and 0 represents nevus. <i>language</i>: Language chosen by the clinician.

本阅片研究分为三个阶段：无AI辅助阶段（阶段1）、AI辅助阶段（阶段2）以及可解释AI（eXplainable AI, XAI）辅助阶段（阶段3）。各阶段生成的数据分别存储于metadata_phase1.csv、metadata_phase2.csv与metadata_phase3.csv文件中。
阶段1共有113名独立参与者。阶段2新增了部分参与者，使得阶段2与阶段3的独立临床医师总数达到116名。
<b>重要提示：</b>每个分组中的第3张与第13张图像完全一致。开展表连接操作时需格外谨慎，重复的图像ID可能会对分析结果造成干扰。在metadata_phase1.csv文件中，每个分组第13张图像对应的AI预测结果为空值，开展分析时请务必考虑该情况。而在metadata_phase2.csv与metadata_phase3.csv文件中，重复图像对应的AI预测结果并未被省略。
<i>participant（参与者）</i>：每位临床医师均分配有唯一参与者ID，该ID由participant列标识。
<i>group（分组）</i>：每位临床医师被随机分配至某一分组，每个分组分配有互不重叠的图像集合。
<i>mask（掩码）</i>：用于图像的内部标识符，可忽略。
<i>benign_malignant（良恶性标签）</i>：真实诊断结果（金标准）。
<i>prediction（医师诊断结果）</i>：临床医师选定的诊断结果，其中1代表黑色素瘤，0代表痣，0.5代表医师诊断为痣但选择切除的情况。
<i>confidence（置信度）</i>：临床医师输入的诊断置信度数值。
<i>trust（信任度）</i>：临床医师输入的信任度数值。
<i>AI_prediction（AI预测结果）</i>：AI生成的诊断预测结果，其中1代表黑色素瘤，0代表痣。
<i>language（语言偏好）</i>：临床医师选择的操作界面语言。