Label-free Femtosecond Laser Imaging Combined With the Fast Lung Artificial Intelligence Model for Rapid Intraoperative Diagnosis of Lung Surgical Specimens: A Multicentre, Prospective, Parallel-workflow, Non-inferiority Study.

This study aims to evaluate whether femtosecond laser imaging combined with FastLung AI model can provide intraoperative diagnostic performance that is non-inferior to standard frozen section diagnosis for pulmonary nodules or suspected pulmonary tumor lesions. Patients scheduled for lung surgery and requiring intraoperative pathological assessment will be prospectively enrolled. After tumor excision, the fresh tumor specimen will be bisected through the central plane. One half will be used for standard frozen section diagnosis, and the mirrored counterpart will be used for femtosecond laser imaging. Both diagnostic results will be compared with the final paraffin-embedded pathological diagnosis as the reference standard. The results of femtosecond laser imaging will not guide intraoperative clinical decision-making.

Sex

All

Minimum age

18 Years

Maximum age

90 Years

Inclusion Criteria:

1. Preoperative imaging suggests a pulmonary nodule, pulmonary space-occupying lesion, or suspected pulmonary tumor lesion.
2. The participant is scheduled to undergo pulmonary wedge resection, segmentectomy, lobectomy, or other pulmonary surgery.
3. Fresh lung tissue specimens can be obtained intraoperatively for femtosecond laser imaging.
4. Intraoperative frozen section diagnosis is planned to assess the nature of the tumor lesion.
5. Corresponding postoperative paraffin-embedded pathological diagnosis can be obtained.
6. The participant or the participant's legally authorized representative has signed the written informed consent form.
7. Patients had not received any antitumor treatment prior to enrollment or specimen collection, including chemotherapy, radiotherapy, targeted therapy, immunotherapy, or other systemic anticancer treatments.

Exclusion Criteria:

1. The intraoperative specimen is insufficient and cannot simultaneously meet the requirements for routine clinical pathological diagnosis and research-related testing.
2. The specimen shows severe carbonization, necrosis, compression, contamination, or improper preservation, and the investigator determines that effective imaging cannot be completed.
3. The femtosecond laser imaging specimen cannot be matched with the corresponding lesion assessed by final paraffin pathology.
4. Final paraffin-embedded pathological diagnosis cannot be obtained.
5. The participant withdraws informed consent.
6. Other conditions that, in the opinion of the investigator, make the participant unsuitable for this study.

• BACKGROUNDXu X, Chung JH, Jheon S, Sung SW, Lee CT, Lee JH, Choe G. The accuracy of frozen section diagnosis of pulmonary nodules: evaluation of inflation method during intraoperative pathology consultation with cryosection. J Thorac Oncol. 2010 Jan;5(1):39-44. doi: 10.1097/JTO.0b013e3181c09f9c. PMID 19934776
• BACKGROUNDHollon TC, Pandian B, Adapa AR, Urias E, Save AV, Khalsa SSS, Eichberg DG, D'Amico RS, Farooq ZU, Lewis S, Petridis PD, Marie T, Shah AH, Garton HJL, Maher CO, Heth JA, McKean EL, Sullivan SE, Hervey-Jumper SL, Patil PG, Thompson BG, Sagher O, McKhann GM 2nd, Komotar RJ, Ivan ME, Snuderl M, Otten ML, Johnson TD, Sisti MB, Bruce JN, Muraszko KM, Trautman J, Freudiger CW, Canoll P, Lee H, Camelo-Piragua S, Orringer DA. Near real-time intraoperative brain tumor diagnosis using stimulated Raman histology and deep neural networks. Nat Med. 2020 Jan;26(1):52-58. doi: 10.1038/s41591-019-0715-9. Epub 2020 Jan 6. PMID 31907460
• BACKGROUNDLan C, Peng Y, Bai M, Zuo H, Li Y, Wu H, Zhang T, Zhu X, He J, Guo D, Chen X, Zhao H, Gao H. Fast multimodal imaging combined with machine learning identifying taurine as a potential marker for breast cancer margin assessment. NPJ Digit Med. 2025 Dec 17;9(1):32. doi: 10.1038/s41746-025-02202-z. PMID 41402452