Precise Position

Precise Position enhances CT patient setup by employing a ceiling-mounted depth camera and a Convolutional Neural Network to automatically determine patient orientation, surview range, and anatomical centering. This AI-driven methodology aims to improve workflow efficiency, positioning accuracy, and inter-operator consistency in CT imaging procedures for patients over 16 years old.

Precise Position is an optional feature designed for Philips Incisive CT systems to assist users in positioning patients prior to CT examinations. Its core function is to reduce patient positioning time and enhance accuracy through an automated workflow. The system employs a ceiling-mounted camera with both color and depth sensing capabilities to capture images of the patient on the CT table. These images are processed by an AI algorithm, specifically a Convolutional Neural Network (CNN), which detects the patient's body joints. This detection enables the system to automatically identify patient orientation, recommend the surview scan range (start/end positions), and determine the center of the patient's anatomy for imaging procedures involving the head, chest, abdomen, pelvis, or combinations of these areas. The system is indicated for patients aged 16 years and older and does not support decubitus orientations; its performance can be affected by patient coverings, incomplete camera view, certain types of clothing (reflective, black, or thick), and the presence of other individuals near the patient. A clinical evaluation of Precise Position involved 80 scan positions (40 with the device and 40 using manual methods), with data assessed by five clinical experts. This evaluation, which did not require radiation for positioning volunteers, aimed to measure improvements in positioning time, vertical (iso-center) accuracy, and surview (horizontal) start/end position consistency. The results indicated that Precise Position can reduce patient positioning time by up to 23%. Furthermore, vertical positioning accuracy was increased by up to 50%, and consistency in both vertical and horizontal positioning among operators improved by up to 70%, suggesting a more standardized and efficient patient setup process for CT scans.

AI Regulatory Concerns

Powered by

Jev reads the full FDA letter and returns calibrated concern probabilities per regulatory category. These are AI estimates, not regulatory advice.

• Q1. What is the reported inter-operator variability in positioning accuracy and time when using Precise Position compared to manual methods across different body regions (head, chest, abdomen, pelvis)?; How does the AI algorithm differentiate between patient anatomy and external objects (e.g., blankets, medical devices attached to the patient) that might be within the camera's field of view?; What are the specific thresholds or conditions (e.g., percentage of body covered, type of reflective material) under which the performance of Precise Position begins to degrade significantly?; Does the system provide real-time feedback to the operator on the confidence level of its automated positioning, and what mechanisms are in place for rapid manual correction?; Were there any specific patient demographic factors (e.g., BMI, presence of metallic implants visible to the camera) beyond age and decubitus position that influenced the system's performance in the clinical evaluation?