Optical differentiation of lung cancer subtypes using laser absorbance, auto-fluorescence emission and Raman spectroscopy: a case series

Introduction

Lung cancer is the leading cause of cancer-related mortality worldwide (1). With the introduction of low-dose computed tomography (LDCT) for lung cancer screening, the detection of small pulmonary nodules, including ground-glass nodules (GGNs), has markedly increased (2). Many of these lesions require surgical biopsy or resection for definitive diagnosis. However, the pulmonary localization for small lung nodule is very challenging and accurate intraoperative pulmonary localization remains a significant unmet need (3). Various localization methods—such as hook-wire placement, dye injection, or radiotracer guidance—are available, all rely on indirect markers rather than direct visualization of the nodule (4). These techniques may cause complications, require additional procedures, and occasionally result in inaccurate resections. At present, few modalities allow real-time identification of pulmonary nodules directly in the surgical field.

Photoacoustic imaging (PAI) has emerged as a promising technology that may assist in pulmonary localization (5,6). In conventional PAI, tissue absorbs laser light and generates acoustic signals through thermoelastic expansion, and signal differences between oxyhemoglobin and deoxyhemoglobin are used to distinguish tissue. Because this mechanism depends on tumor vascularity, hemoglobin-based PAI is not cancer-specific and has limited value for small nodules or GGNs with minimal vascularity (7). To overcome this limitation, it is necessary to establish tumor cell-specific PAI by identifying optical absorbance derived directly from cancer cells rather than hemoglobin. Demonstrating distinct absorbance spectra of lung cancer compared with normal parenchyma could provide the foundation for tumor-specific PAI and enable real-time intraoperative detection of nodules.

Raman spectroscopy is a non-destructive chemical analysis technique and offers complementary potential by providing molecular fingerprints based on inelastic photon scattering (8,9). Unlike PAI, Raman spectroscopy reflects intrinsic biochemical differences between tissues and may reveal subtype-specific features of lung cancer (10-12). Prior reports have shown that phenylalanine and tyrosine bands can discriminate malignant from normal tissues (13) and autofluorescence differences between tumor and normal lung have also been reported (14,15), supporting the use of multimodal optical signatures for tumor characterization.

Therefore, the present study aimed to identify wavelength-specific absorbance differences in optical attenuation between lung cancer and paired normal lung tissue, to evaluate red-excited autofluorescence characteristics, and to investigate Raman spectroscopic variation according to histologic subtype. By characterizing these multimodal optical signatures in ex vivo lung tissue, this study seeks to provide foundational data that may contribute to the future development of optical and photoacoustic techniques for pulmonary nodule localization. We present this article in accordance with the STROBE and AME Case Series reporting checklists (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1941/rc).

Methods

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This prospective study included 12 patients who underwent surgical resection for primary lung cancer without prior neoadjuvant therapy. Informed consent was obtained, and the study was approved by the Institutional Review Board of Chungnam National University Hospital (IRB No. 2019-04-038).

Patients and specimen preparation

Fresh lung specimens were obtained immediately after surgical resection under one-lung ventilation. At the time of specimen collection, the lung tissue was already in a deflated state and was not re-inflated at any point during handling or measurement. Paired samples were collected from tumor tissue and distant normal lung tissue from the same patient. Normal lung tissue was harvested from a site at least 5 cm away from the tumor margin (Figure 1A). Tumor and normal lung tissue were identified based on routine histopathological evaluation by board-certified pathologists, and optical measurements were performed on tissue specimens collected from these pathology-defined regions. Each specimen was trimmed to a standardized size of approximately a diameter of 10 mm with a thickness of 1 mm to minimize variability in optical path length. Tissue thickness was controlled during specimen preparation using a custom slicing tool incorporating a fixed 1 mm spacer between dual blades, enabling reproducible sectioning. Thickness was therefore defined by the mechanical cutting geometry rather than re-measured at individual measurement points. The same slicing procedure was applied to all specimens to ensure consistency across measurements.

Figure 1 Experimental setup and tissue sampling. (A) Gross specimen of resected lung tissue. The yellow circles indicate the cross-sectional area of the tumor from which the tumor sample was obtained. The normal lung tissue sample (red circle) was obtained at a distance of more than 5 cm away from the suspected tumor region (yellow star). (B) Optical attenuation measurement at a fixed wavelength. A continuous-wave diode laser operating at 638 nm was directed onto the tissue sample, and transmitted light was collected through an optical fiber and analyzed using a spectrometer. This configuration enabled quantitative assessment of tissue-level optical attenuation, expressed as OD, for comparison between paired tumor and normal lung tissues. (C) Raman and autofluorescence spectral measurements. The 638 nm laser excitation was delivered to the tissue surface, and scattered and emitted signals were collected via a fiber-optic probe in a backscattering geometry. Optical filters were used to suppress elastically scattered light and isolate Raman scattering and fluorescence emission, which were subsequently analyzed using a spectrometer. This setup enabled acquisition of Raman spectral profiles and autofluorescence emission under identical excitation conditions. *, characteristic Raman peaks/bands used for peak assignment [ν(C-O) of α-helix, tyrosine, and phenylalanine]. OD, optical density.

To reduce blood contamination and minimize hemoglobin-related interference, each specimen was immersed in 20 mL of fresh isotonic saline and gently agitated for 30 seconds. The saline solution was then replaced with fresh saline, and this rinsing procedure was repeated a total of three times. After rinsing, specimens were placed on glass slides for optical measurements. Because measurements were not performed immediately after specimen collection, one to two drops of isotonic normal saline were applied to the tissue surface to prevent dehydration during storage. Samples were kept at room temperature and transferred to the measurement room without additional washing or chemical treatment. Optical measurements were performed with the samples maintained in a moist but non-immersed state and were completed within approximately one hour after specimen collection. For each tissue specimen, optical measurements were acquired at a single representative location. The sample measurement protocol was applied consistently across all tumors and paired normal samples.

Optical measurement system

Preliminary pilot experiments were conducted to explore wavelength-dependent optical attenuation using broadband optical measurements over a wavelength range of approximately 400–1,500 nm. These pilot measurements revealed noticeable differences between tumor and paired normal tissues primarily around 532 and 638 nm, with the largest and most consistent contrast observed at 638 nm. Because broadband measurements were not feasible for all specimens, a fixed-wavelength approach was adopted for the main experiments Accordingly, 638 nm was selected as the representation excitation wavelength for subsequent.

All optical measurements in the main study were performed using a laboratory-built optical spectroscopy system. A single excitation wavelength at 638 nm was employed to evaluate optical attenuation and to enable integration of absorbance, Raman, and red-excited fluorescence measurements within a unified optical system. Raman and fluorescence measurements were not assumed to directly scale with optical absorption; rather, they were used to provide complementary spectroscopic information under identical excitation conditions.

For absorbance measurements, a fixed-wavelength continuous-wave diode laser operating at 638 nm (LP638-SF70, Thorlabs, Newton, NJ, USA) was used as the excitation source. This wavelength was selected based on the pilot study results and because hemoglobin absorption is substantially lower at 638 nm compared with shorter visible wavelengths, thereby reducing the contribution of hemoglobin to the measured optical signal. The laser beam was delivered to the tissue surface via free-space optics, and transmitted light was collected using a fiber-bundle probe and delivered to a spectrometer for spectral acquisition (Figure 1B).

Raman spectra were acquired using a backscattering geometry. The 638 nm excitation light was delivered to the tissue surface through a fiber bundle-based probe, and Raman-scattered light was collected through the same probe and delivered to a spectrometer equipped with a charge-coupled device (CCD) detector. An optical rejection filter (notch filter centered at 638 nm) was used to suppress the excitation wavelength. Autofluorescence measurements were performed using the same optical configuration as the Raman measurements. Tissue specimens were excited with 638 nm diode laser, and the emitted fluorescence was collected through the fiber bundle-based probe in a backscattering geometry (Figure 1C). The same optical geometric and acquisition parameters were applied to all samples to ensure consistency across measurements.

Optical metrics and data processing

The primary optical metric used to characterize tissue optical attenuation in this study was optical density (OD), defined as the logarithmic ratio of incident to transmitted light intensities. The primary signals acquired by the detector were optical intensity measurements. OD was calculated as OD = log10(I₀/I), where I₀ and I denote the incident and transmitted light intensities, respectively. OD represents tissue-level optical attenuation influenced by both absorption and scattering.

Statistical analysis

Continuous variables, including OD values of optical attenuation, were expressed as mean ± standard deviation. Continuous variables between paired lung cancer and normal lung tissues and across histological subtypes were compared using student t-test. A P value <0.05 was considered statistically significant.

Results

Patient characteristics

A total of 12 patients were enrolled in this study, and two tissue samples were obtained from each patient, yielding 24 tumor tissues and 24 normal lung tissues for analysis. The tumor histopathology consisted of eight cases of lung adenocarcinoma and four cases of squamous cell carcinoma. The characteristics of the patients and tumors are summarized in Table 1.

Optical attenuation at 638 nm

At the fixed excitation wavelength of 638 nm, tumor tissues exhibited significantly higher optical attenuation, expressed as OD, compared with paired normal lung tissues (Figure 2). When stratified by histologic subtype, both adenocarcinoma and squamous cell carcinoma demonstrated clear differences in optical attenuation between tumor and corresponding normal lung tissue (Figure 3A,3B). In both subtypes, tumor tissue consistently showed higher OD values than normal lung tissue. Quantitatively, the mean OD values of adenocarcinoma were 2.07±0.15, compared with 1.55±0.10 in paired normal lung tissue. No significant difference in OD values was observed between adenocarcinoma (2.07±0.15) and squamous cell carcinoma (2.02±0.12) (Figure 3C).

Figure 2 Optical attenuation at 638 nm expressed as OD for paired lung tumor and normal lung tissues. Tumor tissues exhibited significantly higher OD values compared with corresponding normal lung tissues, reflecting increased tissue-level optical attenuation arising from combined absorption and scattering. OD, optical density.

Figure 3 Transmission signal profiles and optical attenuation at 638 nm according to histologic subtype. (A) Representative transmission signal profiles of lung adenocarcinoma and paired normal lung tissue at 638 nm, shown as detector output signals (dBm). Representative H&E-stained section is shown (scale bar, 200 µm). These profiles illustrate qualitative differences in transmitted signal between tumor and normal tissues and are presented for representative visualization only. (B) Representative transmission signal profiles of squamous cell carcinoma and paired normal lung tissue at 638 nm, shown as dBm. Representative H&E-stained section is shown (scale bar, 200 µm). (C) Quantitative comparison of optical attenuation at 638 nm expressed as OD. In both adenocarcinoma and squamous cell carcinoma, tumor tissues exhibited significantly higher OD values compared with paired normal lung tissues, whereas no significant difference in OD was observed between adenocarcinoma and squamous cell carcinoma. ADC, adenocarcinoma; OD, optical density; SQCC, squamous cell carcinoma.

Autofluorescence emission spectra

Autofluorescence emission spectra were acquired using 638 nm excitation. In both adenocarcinoma and squamous cell carcinoma, autofluorescence intensity was consistently lower in tumor tissue compared with normal lung tissue. In adenocarcinoma, the peak fluorescence intensity of tumor tissue (5,398.00±579.45) was significantly lower than that of normal lung tissue (8,324.75±379.36) (Figure 4A). Similarly, in squamous cell carcinoma, the peak fluorescence intensity of tumor tissue (3,073.50±1,130.30) was significantly lower than that of normal lung tissue (8,470.13±424.79) (Figure 4B). In addition, the peak fluorescence intensity of adenocarcinoma was significantly higher than that of squamous cell carcinoma (Figure 4C).

Figure 4 Autofluorescence differences between tumor and normal lung tissues and subtype-associated fluorescence characteristics. (A) Fluorescence spectra of ADC and matched normal tissues. Tumor tissues demonstrated a distinct increase in fluorescence intensity compared with normal tissue. (B) Fluorescence spectra of SQCC and matched normal tissues. Tumor tissues demonstrated a distinct increase in fluorescence intensity compared with normal tissue. (C) Comparative analysis of fluorescence intensity between ADC and SQCC tissues. ADC showed significantly higher fluorescence than SQCC, indicating subtype-specific differences in optical properties. ADC, adenocarcinoma; SQCC, squamous cell carcinoma.

Raman spectroscopy

Raman spectra revealed distinct differences between tumor and normal lung tissue. In adenocarcinoma, tumor tissue exhibited relatively stronger peak at 622 cm⁻¹, whereas the 758–785 and 1,320–1,340 cm⁻¹ bands were more prominent in normal lung tissue (Figure 5A). In squamous cell carcinoma, tumor tissue showed relatively stronger peaks at 1,127, 1,173, 1,448, 1,585, and 1,605 cm⁻¹ compared with normal lung tissue (Figure 5B). As in adenocarcinoma, the 758–785 cm⁻¹ band was more prominent in normal lung tissue. Direct comparison between adenocarcinoma and squamous cell carcinoma demonstrated that most Raman bands exhibited stronger intensities in squamous cell carcinoma. However, adenocarcinoma showed relatively stronger peaks in all bands corresponding to phenylalanine and tyrosine (Figure 5C).

Figure 5 Raman spectral distinctions between tumor and normal lung tissues and subtype-specific molecular signatures. (A) Raman spectra of lung ADC and matched normal tissues. Tumor spectra showed higher peak intensities at phenylalanine and tyrosine bands, whereas normal tissues exhibited stronger peaks at nucleic acid and collagen bands. (B) Raman spectra of lung SQCC and matched normal tissues. Tumor spectra showed higher peak intensities at phenylalanine, tyrosine, and lipid bands, whereas normal tissues exhibited stronger intensity at nucleic acid bands. (C) Comparative Raman spectra between ADC and SQCC tissues. Overall signal intensities were higher in SQCC; however, ADC consistently exhibited stronger peaks at phenylalanine and tyrosine bands, highlighting subtype-specific molecular differences. ADC, adenocarcinoma; SQCC, squamous cell carcinoma.

Discussion

In this study, we investigated multimodal optical characteristics of lung adenocarcinoma and squamous cell carcinoma in comparison with paired normal lung tissue using fixed-wavelength optical measurements. Across all modalities examined, tumor tissues demonstrated consistent differences from normal lung tissue, including increased optical attenuation, altered Raman spectral features, and reduced auto fluorescence. These findings indicated the presence of tumor-associated optical signatures at the tissue level that may provide complementary information for intraoperative pulmonary localization (16-19). Optical attenuation analysis revealed significantly higher OD values in both adenocarcinoma and squamous cell carcinoma compared with paired normal lung tissue at 638 nm. Because OD reflects tissue-level optical attenuation influenced by both absorption and scattering, the observed differences should not be interpreted as hemoglobin-free intrinsic absorption. Rather, they likely arise from a combination of factors, including altered cellular density, stromal composition, and residual blood content within tumor tissue. In addition, carbon deposition may also contribute to the attenuation signal (20). Because carbon deposition was not quantitatively assessed in this study, its potential influence cannot be excluded. The absence of a significant difference in OD between adenocarcinoma and squamous cell carcinoma suggests that increased optical attenuation may represent a common feature of malignant lung tissue rather than a histologic subtype-specific marker. Autofluorescence measurements consistently demonstrated reduced fluorescence intensity in tumor tissue relative to normal lung tissue. This finding is consistent with prior reports describing diminished autofluorescence in malignant tissues, which has been attributed to changes in intrinsic fluorophores such as nicotinamide adenine dinucleotide (NADH), flavin adenine dinucleotide (FAD), and collagen-related cross-links (21-23). In addition, adenocarcinoma exhibited higher autofluorescence intensity than squamous cell carcinoma, suggesting that fluorescence measurements may reflect differences in tumor metabolism or stromal remodeling between histologic subtypes (24). While autofluorescence alone is unlikely to provide definitive subtype classification, it may offer complementary contrast for distinguishing tumor from normal lung tissue.

Raman spectroscopy further revealed molecular-level differences between tumor and normal lung tissues. Adenocarcinoma demonstrated relatively stronger Raman bands corresponding to phenylalanine and tyrosine, consistent with increased protein and amino acid content. In contrast, normal lung tissue exhibited stronger peaks at tryptophan/DNA and collagen/nucleic acid-related regions, consistent with preserved extracellular matrix and nucleic acid integrity. Squamous cell carcinoma showed enhanced Raman features in several protein- and lipid-related bands compared with normal lung tissue. Direct comparison between adenocarcinoma and squamous cell carcinoma revealed that, although many Raman bands were more intense in squamous cell carcinoma, adenocarcinoma was characterized by consistently stronger phenylalanine- and tyrosine-associated peaks. These observations suggest that Raman spectroscopy may provide additional discriminatory information beyond simple tumor-normal differentiation (25,26). Taken together, these results demonstrate that lung adenocarcinoma and squamous cell carcinoma share common tumor-associated optical characteristics that distinguish them from normal lung tissue, while also exhibiting modality-specific differences that may reflect underlying histologic and biochemical variations. The integration of optical attenuation, autofluorescence, and Raman spectroscopy highlights the potential value of a multimodal optical approach for characterizing lung cancer tissue in an ex vivo setting. Such complementary optical information may contribute to the development of intraoperative guidance techniques, including photoacoustic and optical imaging approaches, by providing contrast that is not solely dependent on vascular hemoglobin. This study has several limitations that should be acknowledged. First, the sample size was relatively small, and the distribution between adenocarcinoma and squamous cell carcinoma was imbalanced, limiting statistical power for subtype-specific comparisons. Second, all measurements were performed ex vivo, and the optical properties of resected tissue may not fully reflect in vivo conditions due to changes in perfusion, oxygenation, and the tissue microenvironment. Third, although saline rinsing and paired tumor-normal sampling were employed to minimize blood-related effects, residual hemoglobin and tissue scattering cannot be completely excluded. Accordingly, the observed increase in OD at 638 nm should be interpreted as a tissue-level attenuation metric reflecting the combined effects of absorption originating exclusively from cancer cells. In addition, optical measurements were performed at a single excitation wavelength, which precluded analytical separation of hemoglobin contributions using spectral unmixing approaches. Future studies incorporating multi-wavelength or broadband spectroscopy will be necessary to further isolate hemoglobin-independent tissue optical properties. Fourth, although certain Raman bands demonstrated discriminatory potential in this dataset, other commonly reported Raman features were not consistently significant. Finally, histopathological analysis was not performed on the exact same tissue sections following optical measurement; therefore, tissue classification relied on pathology-defined regions rather than section-level confirmation, and the reported optical signals reflect bulk tissue-level responses rather than cell-specific contributions.

Future investigations using larger, balanced patient cohorts and real-time intraoperative measurement systems are warranted to assess the translational potential of multimodal optical spectroscopy. Continued refinement of optical contrast mechanisms and integration into surgical platforms may ultimately support improved localization and margin assessment during lung cancer surgery.

Conclusions

In conclusion, ex vivo multimodal optical analysis of paired lung cancer and normal lung tissues demonstrated reproducible tumor-associated differences in optical attenuation, autofluorescence, and Raman spectral patterns. These optical contrasts reflect tissue-level attenuation and biochemical variation. While further validation in larger cohorts and in vivo settings is required, the present findings support the potential role of multimodal optical spectroscopy as a complementary approach for tumor-associated photoacoustic and spectroscopic guidance in lung cancer surgery.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE and AME Case Series reporting checklists. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1941/rc

Data Sharing Statement: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1941/dss

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1941/prf

Funding: This work was supported by Chungnam National University. No specific grant number was provided.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1941/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. Informed consent was obtained, and the study was approved by the Institutional Review Board of Chungnam National University Hospital (IRB No. 2019-04-038).

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Siegel RL, Giaquinto AN, Jemal A. Cancer statistics, 2024. CA Cancer J Clin 2024;74:12-49. [Crossref] [PubMed]
2. National Lung Screening Trial Research Team. Reduced lung-cancer mortality with low-dose computed tomographic screening. N Engl J Med 2011;365:395-409. [Crossref] [PubMed]
3. Nardini M, Dunning J. Pulmonary nodules precision localization techniques. Future Oncol 2020;16:15-9. [Crossref] [PubMed]
4. Park CH, Han K, Hur J, et al. Comparative Effectiveness and Safety of Preoperative Lung Localization for Pulmonary Nodules: A Systematic Review and Meta-analysis. Chest 2017;151:316-28. [Crossref] [PubMed]
5. Beard P. Biomedical photoacoustic imaging. Interface Focus 2011;1:602-31. [Crossref] [PubMed]
6. Kim C, Cho EC, Chen J, et al. In vivo molecular photoacoustic tomography of melanomas targeted by bioconjugated gold nanocages. ACS Nano 2010;4:4559-64. [Crossref] [PubMed]
7. Yao J, Wang LV. Photoacoustic microscopy. Laser & Photonics Reviews 2013;7:758-78. [Crossref] [PubMed]
8. Shaffer TM, Gambhir SS. Multiplexed Raman Imaging in Tissues and Living Organisms. Methods Mol Biol 2021;2350:331-40. [Crossref] [PubMed]
9. Fales AM, Ilev IK, Pfefer TJ. Evaluation of standardized performance test methods for biomedical Raman spectroscopy. J Biomed Opt 2021;27:074705. [Crossref] [PubMed]
10. Sinica A, Brožáková K, Brůha T, et al. Raman spectroscopic discrimination of normal and cancerous lung tissues. Spectrochim Acta A Mol Biomol Spectrosc 2019;219:257-66. [Crossref] [PubMed]
11. Chen C, Hao J, Hao X, et al. The accuracy of Raman spectroscopy in the diagnosis of lung cancer: a systematic review and meta-analysis. Transl Cancer Res 2021;10:3680-93. [Crossref] [PubMed]
12. Yang X, Wu Z, Ou Q, et al. Diagnosis of Lung Cancer by FTIR Spectroscopy Combined With Raman Spectroscopy Based on Data Fusion and Wavelet Transform. Front Chem 2022;10:810837. [Crossref] [PubMed]
13. Song D, Chen T, Wang S, et al. Study on the biochemical mechanisms of the micro-wave ablation treatment of lung cancer by ex vivo confocal Raman microspectral imaging. Analyst 2020;145:626-35. [Crossref] [PubMed]
14. Hüttenberger D, Gabrecht T, Wagnières G, et al. Autofluorescence detection of tumors in the human lung--spectroscopical measurements in situ, in an in vivo model and in vitro. Photodiagnosis Photodyn Ther 2008;5:139-47. [Crossref] [PubMed]
15. Kilin V, Mas C, Constant S, et al. Health state dependent multiphoton induced autofluorescence in human 3D in vitro lung cancer model. Sci Rep 2017;7:16233. [Crossref] [PubMed]
16. Movasaghi Z, Rehman S, Rehman IU. Raman spectroscopy of biological tissues. Appl Spectrosc Rev 2007;42:493-541.
17. Butler HJ, Ashton L, Bird B, et al. Using Raman spectroscopy to characterize biological materials. Nat Protoc 2016;11:664-87. [Crossref] [PubMed]
18. Auner GW, Koya SK, Huang C, et al. Applications of Raman spectroscopy in cancer diagnosis. Cancer Metastasis Rev 2018;37:691-717. [Crossref] [PubMed]
19. Kendall C, Isabelle M, Bazant-Hegemark F, et al. Vibrational spectroscopy: a clinical tool for cancer diagnostics. Analyst 2009;134:1029-45. [Crossref] [PubMed]
20. Choi M, Shapiro AMJ, Zemp R. Tissue perfusion rate estimation with compression-based photoacoustic-ultrasound imaging J Biomed Opt 2018;23:1-2. (Erratum). [Crossref] [PubMed]
21. Wang M, Long F, Tang F, et al. Autofluorescence imaging and spectroscopy of human lung cancer. Applied Sciences 2017;7:32.
22. Hung J, Lam S, LeRiche JC, et al. Autofluorescence of normal and malignant bronchial tissue. Lasers Surg Med 1991;11:99-105. [Crossref] [PubMed]
23. Croce AC, Bottiroli G. Autofluorescence spectroscopy and imaging: a tool for biomedical research and diagnosis. Eur J Histochem 2014;58:2461. [Crossref] [PubMed]
24. Pesce L, Mastromarino MG, Alì G, et al. Phasor-FLIM and SHG imaging for quantitative analysis of lung cancer autofluorescence. Comput Struct Biotechnol J 2025;30:80-93. [Crossref] [PubMed]
25. Huang Z, McWilliams A, Lui H, et al. Near-infrared Raman spectroscopy for optical diagnosis of lung cancer. Int J Cancer 2003;107:1047-52. [Crossref] [PubMed]
26. Miao Y, Wu L, Qiang J, et al. The application of Raman spectroscopy for the diagnosis and monitoring of lung tumors. Front Bioeng Biotechnol 2024;12:1385552. [Crossref] [PubMed]