RaPID has developed the first computer-aided detection (CADe) system (patent pending) designed to detect retained surgical items (RSIs) on radiologic images. We are using the same technology to develop a computer-aided diagnosis (CADx) system to identify and report the product details of patients' implanted medical devices (IMDs).
Our introductory product is focused on detection of RSIs on X-ray images. Our comprehensive business process modeling involves CADe and specific X-ray imaging sequences that results in the RSIs elimination and more effective usage of the operating room time.
The Clinical Problem
A retained surgical item (RSI) is any surgical sponge or tool inadvertently left behind in a patient’s body in the course of a surgery. Approximately two-thirds of RSIs are surgical sponges, and the other third represents mostly surgical needles, and less frequently surgical instruments.
In the United States, the total number of surgeries in 2013 exceeded 110,000,000 - with an estimated frequency of RSIs between 0.02-1%.
The consequences of RSIs for patients include repeat surgery, permanent injury or even death. Hospitals face the costs of prolonged hospital stays, lawsuits, and loss of hospital credibility.
To prevent RSIs, current patient safety measures include effective communication, mandatory counts of surgical instruments and sponges, methodical wound examinations, and X-ray imaging. Mistakes in counts happen in up to 12.5% of surgeries prompting mandatory X-ray of the surgical field to rule out RSI. In recent years, counts have been enhanced using sponge boards and radio wands.
Even with advances, our current approach to preventing RSIs remains perfectible. 72% to 88% of retained items happen in surgeries with deceptively correct counts. When X-rays are obtained due to a miscount, there may be a delay for the film to be read by a radiologist, and even then objects may be missed due to the relatively low sensitivity of the human eye for RSIs in X-ray images, and a lack of formal training for RSI detection in radiology residency programs.
The RaPID RSI solution integrates CADe software into picture archiving and communication systems (PACSs) data flow to greatly improve the efficiency of X-ray protocols that are mandatory after miscount events or at the end of complex surgeries. Preliminary testing of our beta prototype has resulted in near instantaneous detection of sponges with high accuracy.
Current Product Prototype
RaPID’s patient safety and quality assurance software platform enhances radiologists’ effectiveness in detecting RSIs.
RaPID's CADe/CADx software is hosted on a computer server integrated into the hospital network and configured to receive X-ray images using the DICOM standard. To allow for software support and upgrades, this server is also connected through a firewall into an external RaPID operated infrastructure. Currently the RSI detector software beta prototype is running on the graphic station connected to the initial PACS routing server. If an RSI is detected in the DICOM image, the software embeds a warning sign in the image that appears on the radiologists’ reading screens under PACS.