Back-table specimen scanning using gantry-free hybrid hSPECT/LiDAR imaging: a feasibility study during PSMA-radioguided surgery
Giusi Pisano, Matthias N. van Oosterom, Vera A. Ottens, Anne‐Claire Berrens, Leon J. Slof, Daphne D. D. Rietbergen, Henk G. van der Poel, Pim J. van Leeuwen, Fijs W. B. van Leeuwen
- 发表年份
- 2025
- 引用次数
- 3
- 访问权限
- 开放获取
摘要
Abstract Introduction Prostate-specific membrane antigen (PSMA) targeted precision surgery is becoming increasingly popular. However, the relatively low levels of PSMA-receptor expression and background signal can hinder in vivo lesion detection and margin evaluation. Back-table imaging (ex vivo) potentially provides a means to confirm surgical accuracy. For 99m Tc-PSMA-radioguided surgery, an innovative gantry-free hybrid imaging technique has recently been proposed, namely handheld single-photon emission computed tomography ( h SPECT) combined with light detection and ranging (LiDAR). This study aimed to assess the feasibility and performance of h SPECT/LiDAR in analyzing tissue specimens excised after robotic 99m Tc-PSMA-radioguided surgery. Methods We included samples from 5 prostate cancer patients undergoing primary or salvage robot-assisted resection of 99m Tc-PSMA-I&S avid lesions that were identified using a drop-in gamma probe. 12 samples (1 prostatic tissue, 1 local recurrence tissue, 10 lymph nodes) were analyzed ex vivo using a custom-built specimen tray, including an optical reference tracker for scan registration. LiDAR was used to acquire a surface scan of the specimens, and the 3D OBJ image output was fused with the 3D DICOM of a h SPECT obtained using a handheld gamma camera and DeclipseSPECT tracking system. Results h SPECT/LiDAR imaging provided accurate representation of the 99m Tc-PSMA-I&S uptake within the specimens. In 8 samples, it helped to confirm a true positive lesion. In the remaining 4 samples, non-visualization aligned with negative histopathology (true negative). A strong correlation was found between PSMA- h SPECT/LiDAR and PSMA-PET/CT ( p < 0.05), but no correlation could be established with PSMA-SPECT/CT ( p = 0.515). The count rates fount in the scan correlated to tumor size ( p = 0.016) and were not influenced by the overall specimen’s size ( p = 0.558). Conclusion We present the technical feasibility of a new 3D hybrid modality ( h SPECT/LiDAR) that allows back-table assessment of surgical specimens from the already well validated robotic 99m Tc-PSMA-radioguided surgery workflow.
关键词
相关论文
VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detection
Yin Zhou, Oncel Tuzel
2018
VoxNet: A 3D Convolutional Neural Network for real-time object recognition
Daniel Maturana, Sebastian Scherer
2015
Robots and Jobs: Evidence from US Labor Markets
Daron Acemoğlu, Pascual Restrepo
2019
SECOND: Sparsely Embedded Convolutional Detection
Yan Yan, Yuxing Mao, Bo Li
2018