The Application of Hyperspectral Technology in Skin Detection

長期以來，高光譜成像技術憑借其在農(nóng)作物病蟲害檢測等農(nóng)業(yè)領域的優(yōu)秀表現(xiàn)，展現(xiàn)了其強大的洞察力。然而，高光譜的“慧眼"并非止步于此，它正以其無創(chuàng)、精準的特性，深入洞察人體皮膚的生理指標，為醫(yī)學診斷和健康管理開辟新的篇章。

For a long time, hyperspectral imaging (HSI) technology has demonstrated its powerful insights through its excellent performance in agricultural fields, such as crop pest and disease detection. However, the "all-seeing eye" of hyperspectral technology does not stop there. With its non-invasive and precise characteristics, it is now delving into the physiological indicators of human skin, opening new chapters for medical diagnosis and health management. This technology captures spectral information from objects across different wavelengths, revealing deep physiological changes imperceptible to the human eye.

皮膚灌注與膚色影響

德國的一個研究團隊將高光譜技術用于評估皮膚的灌注情況。它通過檢測組織氧飽和度（StO2）、組織血紅蛋白指數(shù)（THI）、近紅外灌注指數(shù)（NIR-index）和組織水分指數(shù)（TWI）等生理指標，進而判斷皮瓣的血流供應。這些指標的獲取是基于分析皮膚內(nèi)氧合血紅蛋白、脫氧血紅蛋白和黑色素等內(nèi)源性發(fā)色團在不同波長下的吸收和散射特性。

研究發(fā)現(xiàn)，不同膚色和身體部位的皮膚呈現(xiàn)光譜差異。通過分析這些差異并結(jié)合臨床數(shù)據(jù)，研究揭示高光譜在淺膚色人群中灌注評估效果更佳，深膚色受黑色素影響較大，為未來更精準的個性化評估提供了方向。

Skin Perfusion and Skin Tone Influence

A German research team utilized hyperspectral technology to assess skin perfusion. It determines the blood supply to skin flaps by detecting physiological indicators such as Tissue-oxygen-saturation (StO2), Tissue-hemoglobin-index (THI), Near-infrared-perfusion-index (NIR-index), and Tissue-water-index (TWI). The acquisition of these indicators is based on analyzing the absorption and scattering characteristics of endogenous chromophores within the skin, including oxyhemoglobin, deoxyhemoglobin, and melanin, at different wavelengths.

The study found that skin of different tones and body sites exhibits spectral differences. By analyzing these disparities and integrating clinical data, the research revealed that HSI's perfusion assessment is more effective in individuals with lighter skin tones, while darker skin tones are significantly influenced by melanin. This provides direction for more accurate personalized assessments in the future.

黑色素、血紅蛋白 （Hb） 種類和水的吸收光譜

The absorption spectra of melanin, the hemoglobin (Hb) species and water.

4個組織指數(shù)的偽彩色圖像結(jié)果：StO2、NIR指數(shù)、THI和TWI以及高光譜系統(tǒng)的RGB圖像

Results for the false color images of the four tissues indices StO2, NIR-index, THI and TWI with the RGB image as shown by the HSI-device.

血紅蛋白與血氧飽和度

來自歐洲和新西蘭的科研團隊，展示了高光譜在檢測皮膚血紅蛋白含量和血氧飽和度上的應用，并實現(xiàn)其二維可視化。血紅蛋白在特定波長下有明顯吸收峰，不同血氧飽和度也會引起光譜細微變化。研究利用神經(jīng)網(wǎng)絡分析這些光譜特征，精確量化并可視化了血紅蛋白和血氧飽和度。

皮膚的血紅蛋白含量和血氧飽和度是反映人體血液循環(huán)和組織供氧情況的關鍵指標。精確檢測這些指標對于評估貧血、缺氧等血液循環(huán)問題具有重要意義，能為疾病的早期診斷、治療效果監(jiān)測提供客觀依據(jù)。

Hemoglobin and Blood Oxygen Saturation

Research teams from Europe and New Zealand showcased the application of hyperspectral technology in detecting skin hemoglobin content and blood oxygen saturation, enabling their 2D visualization. Hemoglobin exhibits distinct absorption peaks at specific wavelengths, and varying blood oxygen saturation levels lead to subtle spectral changes. The studies leveraged neural networks to analyze these spectral features, accurately quantifying and visualizing hemoglobin and blood oxygen saturation.

Skin hemoglobin content and blood oxygen saturation are crucial indicators reflecting human blood circulation and tissue oxygen supply. Precise detection of these indicators is vital for evaluating circulatory problems such as anemia and hypoxia, providing objective evidence for early disease diagnosis and monitoring treatment efficacy.

該研究的高光譜數(shù)據(jù)處理流程圖

Flowchart of hyperspectral data processing

皮膚色素成分分析

一個歐洲的科研團隊利用了高光譜技術，檢測皮膚中的黑色素和血紅蛋白等色素成分，進而精準測繪皮膚色素分布。不同色素成分在不同波長下有其光譜特性，通過光學模型和反演算法解析高光譜數(shù)據(jù)，可實現(xiàn)色素成分的精確量化和空間分布圖繪制，有助于皮膚疾病診斷、治療效果評估及個性化護膚。

Skin Pigment Component Analysis

A European research team utilized hyperspectral technology to detect pigment components in the skin, such as melanin and hemoglobin, thereby accurately mapping skin pigment distribution. Different pigment components possess unique spectral characteristics at various wavelengths. By analyzing hyperspectral data through optical models and inversion algorithms, precise quantification of pigment components and the creation of spatial distribution maps can be achieved. This contributes to the diagnosis of skin diseases, evaluation of treatment efficacy, and personalized skincare.

利用女性眼周區(qū)域的高光譜圖像獲得的(a)彩色圖像，(b)血容量分數(shù)密度圖，(c)黑色素體積分數(shù)密度圖，(d)血氧飽和圖。

Color image (a), and density images of blood volume fraction(b), melanin volume fraction (c) and oxygen saturation (d) issued from ahyperspectral image of female eye contour.

結(jié)語

高光譜技術，以其優(yōu)勢，正逐步成為醫(yī)學領域的重要工具。它為我們提供了強大的能力，深入理解人體皮膚的復雜生理過程。

我們提供的國產(chǎn)高性能高光譜成像系統(tǒng)，正是邁向這些應用的基礎。憑借優(yōu)秀的性能表現(xiàn)，它能為研究和開發(fā)提供高精度、高質(zhì)量的光譜數(shù)據(jù)。我們專業(yè)的團隊也具備豐富的行業(yè)經(jīng)驗，可以提供專業(yè)的技術建議和支持，幫助客戶實現(xiàn)特定的診斷或分析功能。

Conclusion

Hyperspectral technology, with its unique advantages, is gradually becoming a vital tool in the medical field. It provides us with unprecedented capabilities to deeply understand the complex physiological processes of human skin.

Our domestically produced high-performance hyperspectral imaging systems serve as the foundation for pursuing these applications. With exceptional performance, they provide high-precision, high-quality spectral data for research and development. Our professional team also possesses extensive industry experience, enabling us to offer expert technical advice and support to help clients achieve specific diagnostic or analytical functionalities.

案例來源 /  Source：

1. Pachyn, E. et al. (2024). Investigation on the influence of the skin tone on hyperspectral imaging for free flap surgery. Scientific Reports, 14, 13979.

2. Zherebtsov, E. et al. (2019). Hyperspectral imaging of human skin aided by artificial neural networks. Biomedical Optics Express, 10(7), 3545–3559.

3. Seroul, P. et al. (2016). Model-based Skin Pigment Cartography by High-Resolution Hyperspectral Imaging. Journal of Imaging Science and Technology, 60(6), 060404.