TBNA-protocols Guiding TransBronchial Needle Aspirations Without a Computer in the Operating Room

Abstract

We attempt to maximize the success probability for "blind" biopsies by preoperatively estimating four parameters. These parameters well-define the procedure, given that the bronchoscope resides "somewhere" inside the target branch of the tracheobronchial tree. The calculations are based on a patient specific lung model (CT scan) and a digital model of a flexible bronchoscope. Using both models, we simulate the biopsy and derive frequency distributions for each parameter. Based on these distributions, we choose several parameter sets that together maximize the success probability. The parameters are provided to the surgeon in form of detailed step-by-step instructions (protocol) of how to handle the bronchoscope. To control the parameters in the operating room (OR) we use passive controls. First lung phantom experiments show that we could hit in worst-case 12 mm targets (average-case: 5 mm) with three aspirations.

An “Optimal” k-Needle Placement Strategy Given an Approximate Initial Needle Position

“…There exists a function f : P × T → N , which computes for a given p ∈ P and t ∈ T the necessary needle parameter n ∈ N to hit t from position p. N ⊂ IR 3 is denoted as the "needle parameter domain". Function f () represents a model of the endoscope's active tip deflection, as described in [3]. There also exists a dual functionf : P × N → IR 3 , which computes for a given position p and a needle parameter n the resulting position of the needle tip.…”

“…A typical example for such a procedure is a "transbronchial needle aspiration biopsy" (TBNA) [3], [4]. Traditionally this biopsy is performed by maneuvering a bronchoscope to a suitable site within the tracheobronchial tree.…”

“…Methods to guide TBNAs are based on determining the position and orientation of the bronchoscope's tip. Solomon et al (see [3]) uses position sensors attached to the tip, Mori et al [5] analyzes the video images from a CCD camera inside the bronchoscope's tip to achieve a continuous tracking and our group [3] recently introduced a new approach to the problem by using a model of a flexible bronchoscope, to "predict" the position and orientation of its tip. All three approaches have in common that they determine the tip position approximately, due to sensor, video tracking or model inaccuracies.…”

A Real-Time Deformable Model for Flexible Instruments Inserted into Tubular Structures

“…In recent years numerous attempts have been made to guide endoscopic procedure by determining the position and orientation of the instrument's tip. Solomon et al (see [1]) uses position sensors attached to the tip, Bricault et al [2] introduced the idea of analyzing only the video images and Mori et al [3] significantly improved this approach by achieving continuous tracking independent of the presence of strong features in the image. The authors report a processing time of 6 second per frame.…”

“…The authors report a processing time of 6 second per frame. Our group has recently presented a new approach to the problem [1]. We have suggested the use of a flexible endoscope model to guide a "blind" biopsy (TBNA).…”