多模态人工智能在深肤色患者压力性损伤早期识别与护理中的研究进展

doi:10.3761/j.issn.0254-1769.2026.15.019

中华护理杂志 ›› 2026, Vol. 61 ›› Issue (15): 2155-2160.DOI: 10.3761/j.issn.0254-1769.2026.15.019

• 综述 • 上一篇

多模态人工智能在深肤色患者压力性损伤早期识别与护理中的研究进展

娄洁¹(), 黄蓉蓉²^,^*(), 谢玉生², 刘昭晴², 聂璞², 何可欢², 王浪薰²

¹ 贵州医科大学护理学院护理学院贵阳市 550004
² 贵州医科大学附属医院烧伤整形外科贵阳市 550001

收稿日期:2026-03-09 出版日期:2026-08-10 发布日期:2026-08-04
*通讯作者: 黄蓉蓉，E-mail：2016203030066@whu.edu.cn
作者简介:娄洁：女，本科（硕士在读），护士，E-mail：3073757732@qq.com
第一联系人：作者贡献声明
娄洁：选题、文献搜集、论文撰写；黄蓉蓉：选题指导、论文审阅与修改；谢玉生、刘昭晴、聂璞、何可欢、王浪薰：资料整理与辅助校对
基金资助:
贵州省卫生健康委科学技术基金项目(gzwkj-2024-541)

Progress in the application of multimodal artificial intelligence in early identification and nursing of pressure injuries in dark-skinned populations

LOU Jie¹(), HUANG Rongrong²^,^*(), XIE Yusheng², LIU Zhaoqing², NIE Pu², HE Kehuan², WANG Langxun²

¹ School of Nursing， Guizhou Medical University， Guiyang 550004， China
² Department of Burn and Plastic Surgery， the Affiliated Hospital of Guizhou Medical University， Guiyang 550001， China

Received:2026-03-09 Online:2026-08-10 Published:2026-08-04
* Corresponding author: HUANG Rong-rong，E-mail：2016203030066@whu.edu.cn
Funding program:
Guizhou Provincial Health Commission Science and Technology Foundation Project(gzwkj-2024-541)

摘要/Abstract

摘要：

深肤色患者受先天遗传因素影响，表皮黑色素沉积水平较高，易遮蔽压力性损伤早期红斑体征。多模态人工智能技术通过整合影像学图像、生理信号及临床文本等多源数据，依托数据融合算法挖掘多维度互补信息，有效突破传统单一视觉评估的局限，为深肤色患者压力性损伤的早期精准识别、动态风险预警及个性化护理决策提供了新思路。该文综述深肤色患者压力性损伤的临床评估困境、多模态人工智能的数据融合原理与应用策略及其在护理实践中的研究进展，深入分析该技术临床转化过程中存在的数据偏倚、临床适配不足、隐私伦理约束等关键问题，旨在为构建安全高效的人机协同护理模式、优化压力性损伤防控体系、提升专科护理质量提供参考。

关键词: 多模态, 人工智能, 数据融合, 深肤色, 压力性损伤, 护理, 综述

Abstract:

Affected by congenital genetic factors，individuals with dark skin exhibit high epidermal melanin deposition，which easily obscures early erythema signs of pressure injuries. Multimodal artificial intelligence integrates multi-source data，including medical images，physiological signals and clinical texts. It adopts data fusion algorithms to extract multidimensional complementary information，thereby breaking through the limitations of conventional single visual assessment. It provides objective and innovative insights for the early accurate identification，dynamic risk early warning，and personalized nursing decision-making of pressure injuries in dark-skinned populations. This article reviews the clinical assessment dilemmas of pressure injuries in dark-skinned groups，elaborates on the data fusion principles and application strategies of multimodal artificial intelligence，and summarizes its research progress in nursing practice. Meanwhile，it thoroughly analyzes key challenges in the clinical translation of this technology，such as data bias，insufficient clinical adaptability，and privacy and ethical constraints. The study aims to offer theoretical references for establishing a safe and efficient human-machine collaborative nursing model，optimizing the prevention and control system of pressure injuries，and improving the quality of specialized nursing.

Key words: Multimodality, Artificial Intelligence, Data Fusion, Dark Skin Tones, Pressure Injury, Nursing Care, Review

娄洁, 黄蓉蓉, 谢玉生, 刘昭晴, 聂璞, 何可欢, 王浪薰. 多模态人工智能在深肤色患者压力性损伤早期识别与护理中的研究进展[J]. 中华护理杂志, 2026, 61(15): 2155-2160.

LOU Jie, HUANG Rongrong, XIE Yusheng, LIU Zhaoqing, NIE Pu, HE Kehuan, WANG Langxun. Progress in the application of multimodal artificial intelligence in early identification and nursing of pressure injuries in dark-skinned populations[J]. Chinese Journal of Nursing, 2026, 61(15): 2155-2160.

参考文献 37

[1]	Sugathapala RDUP, Latimer S, Balasuriya A, et al. Prevalence and incidence of pressure injuries among older people living in nursing homes:a systematic review and meta-analysis[J]. Int J Nurs Stud, 2023, 148:104605.
[2]	Oozageer Gunowa N, Oti KA, Jackson D. Early identification of pressure injuries in people with dark skin tones:qualitative perspectives from community-based patients and their carers[J]. J Clin Nurs, 2024, 33(11):4434-4444.
[3]	Johnson J, Johnson AR, Andersen CA, et al. Skin pigmentation impacts the clinical diagnosis of wound infection:imaging of bacterial burden to overcome diagnostic limitations[J]. J Racial Ethn Health Disparities, 2024, 11(2):1045-1055.
[4]	李诏宇. 多模态人工智能正大步走向场景应用新阶段[N]. 科技日报,2022-10-10(6).
	Li ZY. Multimodal artificial intelligence is advancing rapidly into a new phase of scenario-based applications[N]. Sci Tech Daily,2022-10-10(6).
[5]	Schouten D, Nicoletti G, Dille B, et al. Navigating the landscape of multimodal AI in medicine:a scoping review on technical challenges and clinical applications[J]. Med Image Anal, 2025, 105:103621.
[6]	肖爽, 赵庆华, 邹依然, 等. 多模态数据融合的护理信息系统架构及应用分析[J]. 护理学杂志, 2020, 35(19):88-90.
	Xiao S, Zhao QH, Zou YR, et al. Framework of multi-modal data fusion based nursing information system and its appli-cation[J]. J Nurs Sci, 2020, 35(19):88-90.
[7]	Black J, Cox J, Capasso V, et al. Current perspectives on pres-sure injuries in persons with dark skin tones from the natio-nal pressure injury advisory panel[J]. Adv Skin Wound Care, 2023, 36(9):470-480.
[8]	蒋琪霞, 刘国帧, 朱玉玲, 等. Braden量表诊断老年人压力性损伤的适用性和最佳界值的多中心研究[J]. 中华护理杂志, 2023, 58(4):440-445.
	Jiang QX, Liu GZ, Zhu YL, et al. Study on the applicability and optimal cut-off of Braden Scale in the diagnosis of pressure injury in the elderly based on multicenter study data[J]. Chin J Nurs, 2023, 58(4):440-445.
[9]	Cu CW, Dundas NE, Heintz T, et al. Validity of two subjective skin tone scales and its implications on healthcare model fairness[J]. NPJ Digit Med, 2025, 8(1):595.
[10]	Ⅱ Lo H, Hollywood E, Derwin R. Bridging the gap:ICU nurses’ ex-periences in detecting pressure injuries across diverse skin tones[J]. J Tissue Viability, 2025, 34(3):100891.
[11]	Pusey-Reid E, Quinn LW, Wong J, et al. Representation of dark skin tones in foundational nursing textbooks:an image analy-sis[J]. Nurse Educ Today, 2023, 130:105927.
[12]	蒋琪霞, 祝文君, 倪娟, 等. 14家三级甲等医院ICU护士压力性损伤知识现况与影响因素分析[J]. 中国护理管理, 2025, 25(1):64-68.
	Jiang QX, Zhu WJ, Ni J, et al. Current situation and influen-cing factors of pressure injury knowledge among ICU nurses in 14 tertiary grade A hospitals[J]. Chin Nurs Manag, 2025, 25(1):64-68.
[13]	Hill K. Reducing the severity of pressure injuries and iden-tifying opportunities for prevention in a diverse patient popu-lation[J]. Clin Nurse Spec, 2026, 40(1):29-33.
[14]	Sugathapala RDUP, Balasuriya A, Gillespie BM, et al. Eviden-ce-based teaching strategies for assessing pressure injuries in older nursing home residents with darker skin tones[J]. J Nurs Scholarsh, 2025, 57(3):404-411.
[15]	贺敬婷, 应燕萍, 陆秋芳, 等. 多模态数据在护理领域中应用的范围综述[J]. 中华护理杂志, 2026, 61(7):1002-1008.
	He JT, Ying YP, Lu QF, et al. Scoping review on the appli-cation of multimodal data in the nursing field[J]. Chin J Nurs, 2026, 61(7):1002-1008.
[16]	Cai FM, Jiang XQ, Hou XQ, et al. Application of infrared ther-mography in the early warning of pressure injury:a prospec-tive observational study[J]. J Clin Nurs, 2021, 30(3/4):559-571.
[17]	Bates-Jensen BM, Jordan K, Jewell W, et al. Thermal measure-ment of erythema across skin tones:implications for clinical identification of early pressure injury[J]. J Tissue Viability, 2024, 33(4):745-752.
[18]	Ren Q, Wang YJ, Gao WW, et al. Assessing the efficacy of ultrasound deep tissue evaluation in predicting hospital-acqui-red pressure injuries:a diagnostic study[J]. Medicine(Balti-more), 2025, 104(28):e43268.
[19]	周艳, 郭方达, 林晶晶, 等. 智能体压测量床垫的设计和应用[J]. 中华护理杂志, 2023, 58(12):1533-1536.
	Zhou Y, Guo FD, Lin JJ, et al. Design and application of the intelligent body pressure measuring mattress[J]. Chin J Nurs, 2023, 58(12):1533-1536.
[20]	Turmell M, Cooley A, Yap TL, et al. Improving pressure injury prevention by using wearable sensors to cue critical care patient repositioning[J]. Am J Crit Care, 2022, 31(4):295-305.
[21]	Xiang XH, Zhang K, Qin Y, et al. Smart cushions with ma-chine learning-enhanced force sensors for pressure injury risk assessment[J]. ACS Appl Mater Interfaces, 2024, 16(29):38466-38477.
[22]	Osborne Chambers C, Thompson JA. Shedding new light for nurses:enhancing pressure injury prevention across skin tones with sub-epidermal moisture assessment technology[J]. J Adv Nurs, 2024, 80(7):2801-2812.
[23]	Lupiáñez-Pérez I, Gómez-González AJ, Marfil-Gómez RM, et al. Tissue temperature,flux and oxygen of sacral and trochanteric area under pressure of healthy subjects:a quasi-experimental study[J]. J Tissue Viability, 2021, 30(2):207-215.
[24]	Nguyen KA, Patel D, Edalati M, et al. Electronic-medical-re-cord-driven machine learning predictive model for hospital-acquired pressure injuries:development and external valida-tion[J]. J Clin Med, 2025, 14(4):1175.
[25]	Zhang QD, Liu P, Shao PF, et al. Dual-modal optical imaging of tissue perfusion in response to cooling stimulation facili-tates early detection of pressure ulcer[J]. J Biophotonics, 2024, 17(10):e202400188.
[26]	Shaik T, Tao XH, Li L, et al. A survey of multimodal infor-mation fusion for smart healthcare:mapping the journey from data to wisdom[J]. Inf Fusion, 2024, 102:102040.
[27]	Liu TJ, Christian M, Chu YC, et al. A pressure ulcers assess-ment system for diagnosis and decision making using convo-lutional neural networks[J]. J Formos Med Assoc, 2022, 121(11):2227-2236.
[28]	Chen XJ, Liao SF, Wang P, et al. Multimodal data fusionbased risk assessment models and clinical decision support systems for intraoperative acquired pressure injuries[J]. J Surg, 2025, 10:11261.
[29]	Lei CB, Jiang Y, Xu K, et al. Convolutional neural network models for visual classification of pressure ulcer stages:cross-sectional study[J]. JMIR Med Inform, 2025, 13:e62774.
[30]	Yu YZ, Jia HQ, Zhang L, et al. Deep multi-modal skin-ima-ging-based information-switching network for skin lesion reco-gnition[J]. Bioengineering(Basel), 2025, 12(3):282.
[31]	Brehmer A, Seibold C, Egger J, et al. Fine-grained classifica-tion of pressure ulcers and incontinence-associated dermatitis using multimodal deep learning:algorithm development and validation study[J]. JMIR AI, 2025, 4:e67356.
[32]	Shen Y, Li HY, Sun C, et al. Optimizing skin disease diagno-sis:harnessing online community data with contrastive lear-ning and clustering techniques[J]. NPJ Digit Med, 2024, 7(1):28.
[33]	Xu J, Chen DX, Deng XF, et al. Development and validation of a machine learning algorithm-based risk prediction model of pressure injury in the intensive care unit[J]. Int Wound J, 2022, 19(7):1637-1649.
[34]	Dweekat OY, Lam SS, McGrath L. A hybrid system of Braden Scale and machine learning to predict hospital-acquired pres-sure injuries(bedsores):a retrospective observational cohort study[J]. Diagnostics(Basel), 2022, 13(1):31.
[35]	夏冬云, 史婷奇, 陆巍, 等. 压力性损伤临床决策支持系统的研发与应用[J]. 中华护理杂志, 2020, 55(1):50-54.
	Xia DY, Shi TQ, Lu W, et al. Development and application of clinical decision support system for pressure injury[J]. Chin J Nurs, 2020, 55(1):50-54.
[36]	Majjouti K, Priester V, Tapp-Herrenbrueck M, et al. Nursing-centered development of an AI-based decision support system in pressure ulcer and incontinence-associated dermatitis mana-gement:a mixed methods study[J]. BMC Nurs, 2025, 24(1):808.
[37]	Kirkland-Kyhn H, Sengul T, Karadag A, et al. Exploring nurses’ acceptability and readiness for patient-centered artificial inte-lligence systems in pressure injury prevention[J]. Adv Skin Wound Care, 2025, 38(9):488-495.

多模态人工智能在深肤色患者压力性损伤早期识别与护理中的研究进展

Progress in the application of multimodal artificial intelligence in early identification and nursing of pressure injuries in dark-skinned populations

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献 37

相关文章 15

编辑推荐

Metrics

[1]	中华护理学会静脉输液治疗专业委员会, 北京护理学会, （执笔：李佳, 李旭英, 覃惠英, 陈利芬, 吴珍明, 胡泽吟, 罗梦娜, 范育英, 吴嘉慧, 林宇萍, 王蕾, 孙文彦, 冯毕龙, 高伟, 李亚南, 侯罗娅, 李春燕. 中心静脉通路装置拔除护理专家共识[J]. 中华护理杂志, 2026, 61(9): 1157-1162.
[2]	周霞, 马艳艳, 师正坤, 于海霞, 周艳, 胡红玲, 李莜, 陈思思, 张京慧. 隧道式经股静脉PICC置入术在困难静脉通路婴幼儿中的应用研究[J]. 中华护理杂志, 2026, 61(9): 1163-1170.
[3]	黄萍, 李黎, 郭骊莉, 徐莹, 叶赟, 江淑芳. 与无针输液接头相关的导管相关性血流感染风险评估工具的构建与初步应用[J]. 中华护理杂志, 2026, 61(9): 1171-1179.
[4]	李佳, 张玉玲, 邢乐, 张利峰, 牛秀峰, 许莉. 早产儿经上肢置入PICC尖端继发性异位风险预测模型的建立与验证[J]. 中华护理杂志, 2026, 61(9): 1180-1186.
[5]	何娟, 赵蕾蕾, 叶冠军, 余艳芬, 傅晓君, 周琴. 1例双侧无名静脉及上腔静脉人工血管置换术后患者经上肢置入PICC的护理[J]. 中华护理杂志, 2026, 61(9): 1187-1191.
[6]	梁江淑渊, 曾妃, 何鹏, 谢梦珊, 江悦, 蒋宗恒, 王建茗. 静脉-静脉体外膜肺氧合患者早期活动风险等级评估系统的构建及应用研究[J]. 中华护理杂志, 2026, 61(9): 1192-1199.
[7]	陈晨, 顾肖, 郭凡, 王敏, 倪兴梅, 黄琴, 程念开. 3种衰弱评估工具对急诊老年创伤患者不良结局的预测价值[J]. 中华护理杂志, 2026, 61(9): 1200-1207.
[8]	郑煜琳, 刘晋宁, 张静, 郭会敏, 谷艳梅, 张莉莉. 酒精性肝病患者肝移植术后再饮酒现状及其影响因素分析[J]. 中华护理杂志, 2026, 61(9): 1208-1214.
[9]	叶俊霜, 沈丽佳, 孟霞靓, 张玉姣, 蔡根莲, 邵碧云, 周亚辉, 张柳倩, 应金萍. 维持性血液透析患者对口腔衰弱认知和体验的质性研究[J]. 中华护理杂志, 2026, 61(9): 1215-1220.
[10]	王芸姣, 杨玉金, 郑春艳, 彭菲津, 王淑贞, 蒋喜露, 周为民, 王薇, 张娜. 糖尿病下肢动脉硬化闭塞症患者介入术后康复护理方案的构建及应用研究[J]. 中华护理杂志, 2026, 61(9): 1221-1229.
[11]	杨明霞, 安冉, 臧金凤, 李春梅. 老年慢性病共病患者及照护者老老照护能力与自我忽视的相关性分析[J]. 中华护理杂志, 2026, 61(9): 1230-1236.
[12]	梁和静, 许鹊, 史云霞, 祖金美, 黄文静, 杨春旭, 郭明华, 王磊. 颈动脉内膜剥脱术围手术期护理质量评价指标的构建[J]. 中华护理杂志, 2026, 61(9): 1237-1244.
[13]	韩媛媛, 石玉竹, 尚文涵, 么莉. 870所三级甲等医院老年住院患者跌倒伤害发生现状与影响因素分析[J]. 中华护理杂志, 2026, 61(9): 1245-1252.
[14]	夏云芳, 王洪, 崔岱, 付真真, 朱敏. 1例弥漫性毒性甲状腺肿合并超敏反应综合征青少年患者的护理[J]. 中华护理杂志, 2026, 61(9): 1253-1256.
[15]	周益飞, 王海苹, 胡益环, 应莉, 罗松娜. 1例氢氟酸中毒并发呼吸心搏骤停患者的急救护理[J]. 中华护理杂志, 2026, 61(9): 1257-1260.