目的 借助深度聚類在高維深層數(shù)據(jù)處理中的優(yōu)勢，構(gòu)建一種臨床用藥規(guī)律分析模型，挖掘復(fù)雜多維的臨床醫(yī)學(xué)數(shù)據(jù)中的潛在用藥規(guī)律。方法 收集真實(shí)世界臨床腦卒中數(shù)據(jù)集進(jìn)行篩選與規(guī)范，利用圖卷積神經(jīng)網(wǎng)絡(luò)、自動編碼器和自監(jiān)督機(jī)制等技術(shù)，構(gòu)建多源異構(gòu)信息融合的深度聚類模型——多源融合卷積網(wǎng)絡(luò)（multi-source fusion convolutional network，MFCN）。通過2個(gè)數(shù)據(jù)集驗(yàn)證模型的性能，并以腦卒中數(shù)據(jù)集為例，進(jìn)一步分析其臨床用藥規(guī)律，從而驗(yàn)證模型挖掘高維臨床數(shù)據(jù)的有效性。結(jié)果 MFCN模型在DBLP和腦卒中數(shù)據(jù)集上的準(zhǔn)確率分別為79.32%、83.48%，歸一化互信息（normalized mutual information，NMI）分別為0.490 8、0.531 6，調(diào)整蘭德系數(shù)（average rand index，ARI）分別為0.541 9、0.581 7，F(xiàn)1得分（F1-score，F(xiàn)1）分別為0.787 7、0.833 9，其性能指標(biāo)均高于其他模型。在腦卒中數(shù)據(jù)集中，MFCN模型成功挖掘出急性期的中藥組合社團(tuán)，如甘草、茯苓和陳皮等，并發(fā)現(xiàn)“癥-藥”關(guān)聯(lián)組合，如“甘草-脈弦”“甘草-舌紅”“陳皮-肢體乏力”等。結(jié)論 深度聚類能有效處理復(fù)雜高維的臨床數(shù)據(jù)，并揭示臨床用藥的潛在規(guī)律，為臨床經(jīng)驗(yàn)的提取和輔助決策提供了方法學(xué)參考。

Objective To develop a clinical medication pattern analysis model to identify potential medication patterns within multi-dimensional clinical data by utilizing the advantages of deep clustering in high-dimensional and complex data processing. Methods The real-world clinical stroke data were collected for screening and specification. Then, techniques such as graph convolutional neural networks, autoencoders, and self supervised mechanisms are used to construct a deep clustering model—multi-source fusion convolutional network (MFCN) for multi-source information fusion. The performance of the model was validated on two datasets, with the stroke dataset used as a case study to further analyze its clinical medication patterns and demonstrate the effectiveness of the model in mining high-dimensional clinical data. Results The accuracy of the MFCN model on the DBLP and stroke datasets were 79.32% and 83.48%, respectively. The normalized mutual information (NMI) indicators were 0.490 8 and 0.531 6, the average rand index (ARI) indicators were 0.541 9 and 0.581 7, and the F1-score (F1) indicators were 0.787 7 and 0.833 9, respectively. Its performance indicators were higher than those of other models. In the stroke dataset, the MFCN model successfully identified Chinese herbal combinations in the acute phase, such as licorice, Fuling (Poria), and Chenpi (Citri Reticulatae Pericarpium), and discovered “symptom-drug” associations, such as “Gancao (Glycyrrhizae Radix et Rhizoma)-mai xian”, “Glycyrrhizae Radix et Rhizoma-tongue red”, and “Chenpi (Citri Reticulatae Pericarpium)-limb weakness”. Conclusion Deep clustering can effectively handle complex high-dimensional clinical data and reveal potential patterns of clinical medication, with view to providing methodological references for extracting clinical experience and assisting decision-making.

TP18;R285

國家自然科學(xué)基金面上項(xiàng)目（82474352）;湖南省中醫(yī)藥管理局重點(diǎn)項(xiàng)目（A2024011，2023-24）;湖南省自然科學(xué)基金項(xiàng)目（2023JJ60124）;湖南省教育廳科學(xué)研究重點(diǎn)項(xiàng)目（22A0255，22A0281）;湖南省教育廳優(yōu)秀青年項(xiàng)目（22B0400）;湖南省教育廳科學(xué)研究一般項(xiàng)目（22C0195）;長沙市自然科學(xué)基金項(xiàng)目（kq2202265，kq2402172）;湖南中醫(yī)藥大學(xué)校級科研基金項(xiàng)目（2021XJJJ021）