Automatic renal carcinoma biopsy guidance using forward-viewing endoscopic optical coherence tomography and deep learning.

Percutaneous renal biopsy is commonly used for kidney cancer diagnosis. However, the biopsy procedure remains challenging in sampling accuracy. Here we introduce a forward-viewing optical coherence tomography probe for differentiating tumor and normal tissues, aiming at precise biopsy guidance. Totally, ten human kidney samples, nine of which had malignant renal carcinoma and one had benign oncocytoma, were used for system evaluation. Based on their distinct imaging features, carcinoma could be efficiently distinguished from normal renal tissues. Additionally, oncocytoma could be differentiated from carcinoma. We developed convolutional neural networks for tissue recognition. Compared to the conventional attenuation coefficient method, convolutional neural network models provided more accurate carcinoma predictions. These models reached a tissue recognition accuracy of 99.1% on a hold-out set of four kidney samples. Furthermore, they could efficiently distinguish oncocytoma from carcinoma. In conclusion, our convolutional neural network-aided endoscopic imaging platform could enhance carcinoma diagnosis during percutaneous renal biopsy procedures.

Automatic renal carcinoma biopsy guidance using forward-viewing endoscopic optical coherence tomography and deep learning.

Percutaneous renal biopsy (PRB) is commonly used for kidney cancer diagnosis. However, current PRB remains challenging in sampling accuracy. This study introduces a forward-viewing optical coherence tomography (OCT) probe for differentiating tumor and normal tissues, aiming at precise PRB guidance. Five human kidneys and renal carcinoma samples were used to evaluate the performance of our probe. Based on their distinct OCT imaging features, tumor and normal renal tissues can be accurately distinguished. We examined the attenuation coefficient for tissue classification and achieved 98.19% tumor recognition accuracy, but underperformed for distinguishing normal tissues. We further developed convolutional neural networks (CNN) and evaluated two CNN architectures: ResNet50 and InceptionV3, yielding 99.51% and 99.48% accuracies for tumor recognition, and over 98.90% for normal tissues recognition. In conclusion, combining OCT and CNN significantly enhanced the PRB guidance, offering a promising guidance technology for improved kidney cancer diagnosis.

Ultrasound-Guided Needle Biopsy of Neck Lymph Nodes in Patients With Suspected Lung Cancer: Are the Specimens Sufficient for Complete Pathologic Evaluation to Guide Patient Management?

BACKGROUND: The purpose of this study is to determine the ability of ultrasound guided needle biopsy of a neck lymph node to provide adequate tissue for complete pathologic evaluation of suspected metastatic lung cancer, including molecular testing for epidermal growth factor receptor gene mutations by pyrosequencing and anaplastic lymphoma kinase gene rearrangement by fluorescence in situ hybridization. METHODS: Institutional review board approval was obtained and the requirement for informed consent was waived. All ultrasound guided neck biopsies performed July 1, 2011, to June 30, 2015, were retrospectively reviewed, and all biopsies performed for suspected lung cancer metastatic to supraclavicular and cervical lymph nodes were included. RESULTS: Forty patients with suspected lung cancer underwent ultrasound-guided needle biopsy of an abnormal appearing neck lymph node identified on preprocedure computed tomography or positron emission tomography/computed tomography. Thirty-seven patients were subsequently diagnosed with lung cancer and 3 were diagnosed with lymphoma. A definitive pathologic diagnosis was rendered in 95% of neck node biopsies (38/40; 95% confidence interval, 84%-99%). Of the 36 specimens diagnostic for lung cancer, 16 were considered for further molecular testing and the specimen was adequate for molecular testing in 15 (94%; 73%-100%) cases. Therefore, the neck node biopsy specimens were adequate for complete pathologic workup in 93% (37/40; 81%-98%). No complications related to the biopsies were observed. CONCLUSIONS: In patients presenting with suspected lung cancer and suspicious neck lymph nodes, ultrasound-guided needle biopsy frequently provides adequate tissue for complete pathologic evaluation and eliminates the need for more invasive procedures.

Ultrasound Versus Computed Tomographic Guidance for Percutaneous Biopsy of Chest Lesions.

OBJECTIVES: To compare ultrasound (US) versus computed tomography (CT) for primary guidance during needle biopsy of chest lesions. METHODS: Institutional Review Board approval was obtained for this Health Insurance Portability and Accountability Act-compliant retrospective study, and the need for informed consent was waived. All US- and CT-guided chest biopsy procedures performed between January 1, 2012, and October 15, 2014, at our institution were reviewed, and all procedures targeting peripheral intrathoracic and chest wall lesions were included. Axillary lesions, lung lesions without peripheral pleural contact, and mediastinal lesions without a transcutaneous US window were excluded. Radiologic, pathologic, and clinical records were reviewed. RESULTS: Fifty-five procedures with primary US guidance (23 lung, 6 pleural, 2 mediastinal, and 24 chest wall) and 130 CT procedures (88 lung, 10 pleural, 7 mediastinal, and 25 chest wall) were performed. Diagnostic samples were obtained in 98% (54 of 55) of US procedures and 87% (113 of 130) of CT procedures (P = .02). Pneumothorax requiring treatment occurred in 2% (1 of 55) of US procedures and 5% (7 of 130) of CT procedures (P = .25). Computed tomographic localization was used in 29% (16 of 55) of US procedures. Nevertheless, the average patient radiation dose was significantly less in US procedures (182 mGy-cm) versus CT procedures (718 mGy-cm; P< .01). The average procedure time was 40 minutes for US and 38 minutes for CT (P = .39). The average lesion size was 4.5 cm for US and 4.9 cm for CT (P = .14). CONCLUSIONS: During biopsy of peripheral intrathoracic lesions and chest wall lesions, primary US guidance resulted in a higher likelihood of a diagnostic sample and a decreased patient radiation dose compared with CT guidance.