Faculty Bibliography

Purpose: To assess the safety of percutaneous cryoablation for treatment of pulmonary metastases < 3.5 cm. Material and methods: Patients with up to 6 pulmonary metastases were treated in a multicenter, prospective study with CT-guided percutaneous cryoablation. Patient follow-up was performed at 1 week and 3, 6, 12, and 24 months. Complications were assessed using the CTCAE 4.03. Results: Two-hundred nineteen tumors in 126 patients (64 males, 62 females; mean age 64 years) were treated over 163 procedures, constituting the largest prospective multicenter lung cryoablation trial to date. Sixty-two percent (n=78) of the patients had 1 tumor, and 40% had 2 or more. Mean tumor size was 1.2 cm (0.4- 4.5 cm). General anesthesia was employed in 69% (n=113) of procedures, conscious sedation in 29% (n=47), and local anesthesia in 2% (n=3). Treatment time ranged from 30 to 225 min. (mean=72.9 min). Ninety-eight procedure-related events occurred within 30 days of the treatment. Thirty-seven percent were classified CTCAE grade 1 (n=36), 57% grade 2 (n=57), 5% grade 3 (n=5), and 1% grade 4 (n=1). The most frequent event was pneumothorax, of which 8% (n=4) were grade 1, 83% (n=40) grade 2, and 8% (n=4) grade 3. Six percent (n=6) of all events were grade 3 or higher: grade 3 pneumothoraces (n=4), grade 3 pleural hemorrhage (n=1), and grade 4 gas embolism (n=1). All events resolved without further sequelae. Conclusion: Percutaneous CT-guided cryoablation demonstrates an acceptable safety profile for treatment of lung metastases < 3.5 cm

Markov random field (MRF) model has been widely employed in edge-preserving regional noise smoothing penalty to reconstruct piece-wise smooth images in the presence of noise, such as in low-dose computed tomography (LdCT). While it preserves edge sharpness, its regional smoothing may sacrifice tissue image textures, which have been recognized as useful imaging biomarkers, and thus it may compromise clinical tasks such as differentiating malignant vs. benign lesions, e.g., lung nodules or colon polyps. This study aims to shift the edgepreserving regional noise smoothing paradigm to texturepreserving framework for LdCT image reconstruction while retaining the advantage of MRF's neighborhood system on edge preservation. Specifically, we adapted the MRF model to incorporate the image textures of muscle, fat, bone, lung, etc. from previous full-dose CT (FdCT) scan as a priori knowledge for texture-preserving Bayesian reconstruction of current LdCT images. To show the feasibility of the proposed reconstruction framework, experiments using clinical patient scans were conducted. The experimental outcomes showed a dramatic gain by the a priori knowledge for LdCT image reconstruction using the commonly-used Haralick texture measures. Thus, it is conjectured that the texture-preserving LdCT reconstruction has advantages over the edge-preserving regional smoothing paradigm for texture-specific clinical applications.