摘要   隨著環(huán)境和健康研究數(shù)據(jù)共享及可用性的不斷提升，涉及環(huán)境與人體健康的數(shù)據(jù)集數(shù)量急劇增加。然而，這些環(huán)境健康大型數(shù)據(jù)集多樣且復(fù)雜，傳統(tǒng)的流行病學(xué)和環(huán)境健康模型難以有效分析，因此催生了一個(gè)環(huán)境健康研究的新手段。人工智能（AI）技術(shù)在環(huán)境健康領(lǐng)域的應(yīng)用正迅速發(fā)展，為新污染物篩選和毒性預(yù)測(cè)、生物監(jiān)測(cè)、風(fēng)險(xiǎn)評(píng)估和健康保護(hù)提供了新穎且強(qiáng)大的工具。其中，先進(jìn)的機(jī)器學(xué)習(xí)（ML）算法能夠揭示人類難以察覺的規(guī)律，在生物標(biāo)志物識(shí)別、疾病預(yù)防和環(huán)境工程優(yōu)化等方面表現(xiàn)出重要潛力，為環(huán)境健康研究和技術(shù)創(chuàng)新提供新的思路和突破口。然而，ML技術(shù)在環(huán)境健康領(lǐng)域的應(yīng)用仍面臨數(shù)據(jù)質(zhì)量、模型解釋性以及跨學(xué)科合作等挑戰(zhàn)。本文將綜述ML技術(shù)在環(huán)境健康領(lǐng)域的最新應(yīng)用進(jìn)展，探討其優(yōu)勢(shì)、挑戰(zhàn)以及未來的發(fā)展方向，以期為環(huán)境保護(hù)和公共健康領(lǐng)域的研究和實(shí)踐提供有價(jià)值的參考。

Abstract   As the data sharing and availability in environmental and health research continue to improve, the number of large datasets for environmental and human health has increased dramatically. However, these large environmental health datasets are diverse and complex, and traditional epidemiological and environmental health models are difficult to effectively analyze, leading to the development of a new approach to environmental health research. The application of artificial intelligence (AI) technology in environmental health is rapidly developing, providing novel and powerful tools for new pollutant screening and toxicity prediction, biomonitoring, risk assessment, and health protection. Among them, advanced machine learning (ML) algorithms can reveal laws that are difficult for humans to detect, showing important potential in biomarker identification, disease prevention, and environmental engineering optimization. This can provide new ideas and breakthroughs for environmental health research and technological innovation. However, the application of ML technology in the field of environmental health still faces challenges such as data quality, model interpretability, and interdisciplinary cooperation. This paper will review the latest progress in the application of ML technology in the field of environmental health, discuss its advantages, challenges, and future development directions, with the aim of providing valuable references for research and practice in the fields of environmental protection and public health.

[1] 江桂斌, 宋茂勇. 環(huán)境暴露與健康效應(yīng)[M]. 北京：科學(xué)出版社, 2020. [2] 劉靜怡, 孟聰申, 韓京秀. 1990—2019年全球環(huán)境危險(xiǎn)因素疾病負(fù)擔(dān)—GBD2019數(shù)據(jù)再分析[J]. 環(huán)境衛(wèi)生學(xué)雜志, 2023, 13(3): 170-176. [3] 支夢(mèng)雪, 王建設(shè). 暴露組學(xué)在識(shí)別環(huán)境污染物及其健康危害中的應(yīng)用進(jìn)展[J]. 色譜, 2024, 42(2): 142-149. [4] 李立明, 王波, 呂筠, 等. 我國(guó)公共衛(wèi)生科技創(chuàng)新的現(xiàn)狀與挑戰(zhàn)[J]. 中國(guó)科學(xué)基金, 2024, 38(2): 303-307. [5]

HANDELMAN G S, KOK H K, CHANDRA R V, et al. eD octor: machine learning and the future of medicine[J]. Journal of Internal Medicine, 2018, 284(6): 603-619.

[6]

LIU B, DING M, SHAHAM S, et al. When machine learning meets privacy: a survey and outlook[J]. ACM computing surveys, 2022, 54(2): 1-36.

[7]

DONG S, WANG P, ABBAS K. A survey on deep learning and its applications[J]. Computer science review, 2021, 40: 100379.

[8]

ALZUBAIDI L, ZHANG J, HUMAIDI A J, et al. Review of deep learning: concepts, CNN architectures, challenges, applications, future directions[J]. Journal of big data, 2021, 8(1): 53.

[9] 鄭玉新. 暴露評(píng)估與暴露組研究——探索環(huán)境與健康的重要基礎(chǔ)[J]. 中華預(yù)防醫(yī)學(xué)雜志, 2013, 47(2): 99-100. [10] 孟甜, 曹瑩, 劉曉雪, 等. 環(huán)境應(yīng)急監(jiān)測(cè)技術(shù)研究進(jìn)展與展望[J]. 環(huán)境保護(hù), 2023, 14(51): 34-39. [11]

MAGI E, DI CARRO M. Marine environment pollution: The contribution of mass spectrometry to the study of seawater[J].Mass spectrometry reviews,2018, 37(4): 492-512.

[12] 李艷, 吳欣宜, 王全龍, 等. 機(jī)器學(xué)習(xí)輔助光譜分析技術(shù)在環(huán)境微/納塑料研究中的應(yīng)用[J]. 中國(guó)無機(jī)分析化學(xué), 2024, 14(8): 1137-1146. [13]

ZHAO Q M, YU Y, GAO Y C, et al. Machine learningbased models with high accuracy and broad applicability domains for screening PMT/vPvM substances[J]. Environmental Science & Technology, 2022, 56(24): 17880- 17889.

[14]

SUN X, ZHANG X, MUIR D C G, et al. Identification of potential PBT/POP-like chemicals by a deep learning approach based on 2D structural features[J]. Environmental science & technology, 2020, 54(13): 8221-8231.

[15]

TIAN X, BEéN F, B?UERLEIN P S. Quantum cascade laser imaging (LDIR) and machine learning for the identification of environmentally exposed microplastics and polymers[J]. Environmental research, 2022, 212: 113569.

[16]

LI R Y, YAN C Q, MENG Q P, et al. Key toxic components and sources affecting oxidative potential of atmospheric particulate matter using interpretable machine learning: Insights from fog episodes[J]. Journal of hazardous materials, 2024, 465: 133175.

[17]

DENG F C, QIN G Q, CHEN Y Y, et al. Multi-omics reveals 2-bromo-4,6-dinitroaniline (BDNA)-induced hepatotoxicity and the role of the gut-liver axis in rats[J]. Journal of hazardous materials, 2023, 457: 131760.

[18]

LI Z L, WANG J, YUE H, et al. Applying metabolic modeling and multi-omics to elucidate the biotransformation mechanisms of marine algal toxin domoic acid (DA) in sediments[J]. Journal of hazardous materials, 2024, 472:134541.

[19]

PARK Y J, RAHMAN M S, PANG W K, et al. Systematic multi-omics reveals the overactivation of T cell receptor signaling in immune system following bisphenol A exposure[J]. Environmental pollution, 2022, 308: 119590.

[20]

REEL P S, REEL S, PEARSON E, et al. Using machine learning approaches for multi-omics data analysis: A review[J]. Biotechnology advances, 2021, 49: 107739.

[21]

LI R F, LI L X, XU Y G, et al. Machine learning meets omics: applications and perspectives[J]. Briefings in Bioinformatics, 2022, 23(1): bbab460.

[22]

GUO W J, LIU J, DONG F, et al. Review of machine learning and deep learning models for toxicity prediction [J]. Experimental biology and medicine, 2023: 15353702231209421.

[23]

PENG T, WEI C H, YU F B, et al. Predicting nanotoxicity by an integrated machine learning and metabolomics approach[J]. Environmental pollution, 2020, 267: 115434.

[24]

SU Z K, ZHANG Y L, HONG S Y, et al. Immune regulation patterns in response to environmental pollutant chromate exposure-related genetic damage: a cross-sectional study applying machine learning methods[J]. Environmental science & technology, 2024, 58(17): 7279-7290.

[25]

LUAN H. Machine learning for screening active metabolites with metabolomics in environmental science[J]. Environmental science: advances, 2022, 1(5): 605-611.

[26]

LI L, CHING W K, LIU Z P. Robust biomarker screening from gene expression data by stable machine learningrecursive feature elimination methods[J]. Computational biology and chemistry, 2022, 100: 107747.

[27]

TORUN F M, VIRREIRA WINTER S, DOLL S, et al. Transparent exploration of machine learning for biomarker discovery from proteomics and omics data[J]. Journal of proteome research, 2023, 22(2): 359-367.

[28]

WEI X R, TANG X W, LIU N, et al. PyCoCa: A quantifying tool of carbon content in airway macrophage for assessment the internal dose of particles[J]. Science of the total environment, 2022, 851(Part 1): 158103.

[29] 胡貴平, 陳章健, 唐仕川, 等. 生物監(jiān)測(cè)在暴露組評(píng)價(jià)中的應(yīng)用[J]. 中華預(yù)防醫(yī)學(xué)雜志, 2018, 52(2): 945-948. [30]

LIU X, LU D W, ZHANG A Q, et al. Data-driven machine learning in environmental pollution: gains and problems [J]. Environmental science & technology, 2022, 56(4): 2124-2133.

[31] 姚絮. 環(huán)境重金屬聯(lián)合暴露評(píng)價(jià)及其對(duì)重要健康結(jié)局的影響[D]. 合肥：安徽醫(yī)科大學(xué), 2023. [32] 張俊鴻, 閆馨, 張俊. 空氣污染與人體健康的關(guān)系[J]. 山西醫(yī)藥雜志, 2021, 50(24): 3339-3341. [33]

BOBB J F, VALERI L, CLAUS H B, et al. Bayesian kernel machine regression for estimating the health effects of multi-pollutant mixtures[J]. Biostatistics, 2015, 16(3): 493-508.

[34]

MéNDEZ M,MERAYO M G, Nú?EZ M. Machine learning algorithms to forecast air quality: a survey[J]. Artificial intelligence review, 2023, 56(9): 10031-10066.

[35]

DU S D, LI T R, YANG Y, et al. Deep air quality forecasting using hybrid deep learning framework[J]. IEEE transactions on knowledge and data engineering, 2021, 33(6): 2412-2424.

[36]

MA J, CHENG J C P, LIN C, et al. Improving air quality prediction accuracy at larger temporal resolutions using deep learning and transfer learning techniques[J]. Atmospheric environment, 2019, 214: 116885.

[37] 張宇, 萬爽, 向準(zhǔn), 等. 基于環(huán)境氣象因素的機(jī)器學(xué)習(xí)模型預(yù)測(cè)呼吸系統(tǒng)疾病急診量[J]. 中國(guó)數(shù)字醫(yī)學(xué), 2023, 18(7): 40-45. [38]

KONTOS Y N, KASSANDROS T, PERIFANOS K, et al. Machine learning for groundwater pollution source identification and monitoring network optimization[J]. Neural computing and applications, 2022, 34(22): 19515-19545.

相關(guān)知識(shí)

FESE ?朱彤院士：中國(guó)環(huán)境健康研究展望—論文—科學(xué)網(wǎng)
《3—6歲兒童學(xué)習(xí)與發(fā)展指南》中健康領(lǐng)域的目標(biāo)解讀
走進(jìn)健康4.0：醫(yī)療健康領(lǐng)域的數(shù)字化革新之路
深耕健康領(lǐng)域 三星以先進(jìn)技術(shù)開啟智能穿戴健康新時(shí)代
用于健康與健身應(yīng)用的傳感器技術(shù)
北京大學(xué)公共衛(wèi)生學(xué)院成功舉辦“環(huán)境健康與美好生活”學(xué)術(shù)論壇
熱烈慶祝北京大學(xué)“環(huán)境健康系”成立!
尖端技術(shù)+臨床實(shí)踐=堪培拉大學(xué)醫(yī)學(xué)健康學(xué)院
桂陽縣衛(wèi)生健康局構(gòu)建醫(yī)療服務(wù)領(lǐng)域信用聯(lián)合獎(jiǎng)懲機(jī)制，優(yōu)化營(yíng)商環(huán)境
智慧賦能健康 科技引領(lǐng)發(fā)展

網(wǎng)址: 機(jī)器學(xué)習(xí)技術(shù)在環(huán)境健康領(lǐng)域中的應(yīng)用進(jìn)展 http://m.u1s5d6.cn/newsview30758.html