US-12620458-B2 - Methods, systems, and computer readable media for automated assessment of aseptic technique of compounding in a compounding hood

US12620458B2US 12620458 B2US12620458 B2US 12620458B2US-12620458-B2

Abstract

A method for automated assessment of aseptic technique of compounding in a compounding hood includes collecting, using sensors positioned in or around a compounding hood, data from which positions, orientations, and movements of objects used in an aseptic compounding task can be determined. The method further includes feeding the data into an automated aseptic technique evaluator. The method further includes identifying, using the automated aseptic technique evaluator and from the data, phases of the aseptic compounding task. The method further includes automatically detecting, by the automated aseptic technique evaluator, errors occurring during at least some of the phases. The method further includes generating and displaying, by the automated aseptic technique evaluator, output indicative of the errors.

Inventors

Stephen Frederick Eckel
Robert Charles Hubal
Adam Charles Kiefer
Ryan Patrick MacPherson
Colin Thomas Cabelka

Assignees

THE UNIVERSITY OF NORTH CAROLINA AT CHAPEL HILL

Dates

Publication Date: 20260505
Application Date: 20250125

Claims (20)

1 . A method for automated assessment of aseptic technique of compounding in a compounding hood, the method comprising: collecting, using sensors positioned in or around a compounding hood, data from which positions, orientations, and movements of objects used in an aseptic compounding task can be determined; feeding the data into an automated aseptic technique evaluator; identifying, using the automated aseptic technique evaluator and from the data, phases of the aseptic compounding task; automatically detecting, by the automated aseptic technique evaluator, errors occurring during at least some of the phases, wherein automatically detecting the errors includes mapping the positions of objects into a three-dimensional coordinate system, determining, from the positions of the objects, a path taken by the object in the three-dimensional coordinate system, quantitatively defining a three-dimensional region in the three-dimensional coordinate system indicating a path tolerance for aseptic technique in the three-dimensional coordinate system, and determining non-compliance with aseptic technique when the path taken is outside of the quantitatively defined three-dimensional region; and generating and displaying, by the automated aseptic technique evaluator, output indicative of the errors.
2 . The method of claim 1 wherein collecting the data using the sensors positioned in or around the compounding hood includes collecting the data using cameras or other optical or other types of sensors positioned in or around the compounding hood.
3 . The method of claim 1 wherein collecting the data using the sensors positioned in or around the compounding hood includes collecting the data using Internet of Things (IoT) sensors located on compounding instruments or compounding materials.
4 . The method of claim 1 wherein identifying the phases includes applying a computer-vision-based or equivalent sensor-based object detection model to identify the objects.
5 . The method of claim 1 wherein identifying the phases includes using a computer-vision-based or equivalent sensor-based pose estimation model to create time series data representing positions and orientations of the objects at different times.
6 . The method of claim 5 wherein identifying the phases includes utilizing task phase identification heuristics to identifying each phase.
7 . The method of claim 1 wherein identifying the phases includes identifying features of a user's gloved hands from the data collected from the sensors and tracking position, orientation, and movement of the identified features in different frames of the data.
8 . The method of claim 1 wherein automatically detecting the errors includes applying a classification model that generates, for each phase, a probability score indicative of a likelihood of occurrence of an error and determining that the error has occurred when the score exceeds a threshold value.
9 . The method of claim 1 wherein generating and displaying the output includes generating and displaying a dashboard interface that indicates the errors that occurred during performance of the aseptic compounding task.
10 . The method of claim 9 wherein the dashboard interface displays in real time during performance of the aseptic compounding task best practice advisory actions for a student/user to perform during the aseptic compounding task, or in batch after performance of aseptic compounding tasks assessment of performance during the aseptic compounding tasks.
11 . The method of claim 1 wherein the automated aseptic technique evaluator is implemented using at least one trained machine learning classifier.
12 . A system for automated assessment of aseptic technique of compounding in a compounding hood, the system comprising: at least one processor; a plurality of sensors positionable in or around a compounding hood for collecting data from which positions, orientations, and movements of objects used in an aseptic compounding task can be determined; and an automated aseptic technique evaluator implemented using the at least one processor for receiving the data, identifying from the data phases of the aseptic compounding task, automatically detecting errors occurring during at least some of the phases, and generating and displaying output indicative of the errors, wherein automatically detecting the errors includes mapping the positions of objects into a three-dimensional coordinate system, determining, from the positions of the objects, a path taken by the object in the three-dimensional coordinate system, quantitatively defining a three-dimensional region in the three-dimensional coordinate system indicating a path tolerance for aseptic technique in the three-dimensional coordinate system, and determining non-compliance with aseptic technique when the path taken is outside of the quantitatively defined three-dimensional region.
13 . The system of claim 12 wherein the sensors comprise cameras or other optical or other types of sensors positionable in or around the compounding hood.
14 . The system of claim 12 wherein the sensors comprise Internet of Things (IoT) sensors locatable on compounding instruments or compounding materials.
15 . The system of claim 12 wherein the automated aseptic technique evaluator is configured to use a computer-vision-based or equivalent sensor-based object detection model to identify the objects.
16 . The system of claim 15 wherein the automated aseptic technique evaluator is configured to use a computer-vision-based or equivalent sensor-based pose estimation model to create time series data representing positions and orientations of the objects at different times.
17 . The system of claim 16 wherein the automated aseptic technique evaluator is configured to utilize task phase identification heuristics to identify each phase.
18 . The system of claim 12 wherein the automated aseptic technique evaluator is configured to identify the phases by identifying features of a user's gloved hands from the data collected from the sensors and tracking position, orientation, and movement of the identified features in different frames of the data.
19 . The system of claim 12 wherein the automated aseptic technique evaluator is configured to detect the errors by applying a classification model that generates, for each phase, a probability score indicative of a likelihood of occurrence of an error and determining that the error has occurred when the score exceeds a threshold value.
20 . The system of claim 12 wherein the automated aseptic technique evaluator is configured to generate a dashboard interface that indicates the errors that occurred during the performance of the aseptic compounding task.

Description

PRIORITY CLAIM This application is a continuation of PCT Patent Application No. PCT/US2023/029079, filed Jul. 31, 2023, which claims the priority benefit of U.S. Provisional Patent Application Ser. No. 63/393,766, filed Jul. 29, 2022, the disclosure of each of which is incorporated herein by reference in its entirety. TECHNICAL FIELD The subject matter described herein relates to tracking aseptic technique during compounding in a compounding hood, defined as a biological safety cabinet, a laminar airflow workbench, or other similar device. More particularly, the subject matter described herein relates to automated tracking and characterizing of aseptic technique of compounding in a compounding hood. BACKGROUND In 2012, a tragedy occurred in sterile compounding that garnered national attention. The New England Compounding Center (NECC) had compounded sterile methylprednisolone injections that were distributed to 23 different states. Unfortunately, these were contaminated with a fungus that led to meningitis in patients receiving the compounded product, resulting in over 800 individuals being sickened and 76 deaths.1 Besides this alarming event, many other compounding tragedies have occurred but not received the same attention.2-9 Since the NECC meningitis outbreak in 2012, the FDA has increased its oversight of compounding pharmacies. Before that event, these businesses were monitored by state boards of pharmacy and were infrequently inspected against United States Pharmacopeia guidance documents. In 2013, the Drug Quality and Security Act was passed and gave more authority to the FDA to oversee compounding. This amended the Federal Food, Drug, and Cosmetic Act to create two categories of compounded products based upon risk level for the patient: compounding pharmacies (section 503a) and outsourcing facilities (section 503b). Since this became law, understanding the various regulations, preparing the pharmacy for compliance, and educating employees on appropriate aseptic technique has been a major focus for all employers. This includes both hospitals and community pharmacies. Despite this increased attention, however, there continues to be issues with compounded products as documented by the FDA through inspections. A warning letter (Form FDA 483) posted on Feb. 5, 2019 to a compounding pharmacy indicated that FDA inspectors found serious deficiencies in preparing sterile drug products. Some of the violations that were noted from this inspection include:10 Personnel were observed conducting aseptic manipulations that blocked the movement of first pass air over and around open vials.The media fills were not performed under the most challenging or stressful conditions.Personnel engaged in aseptic processing were observed with exposed hands and wrists in the ISO 5 area.The inspectors noted a “lack of assurance” in aseptically producing drug products within the facility. To prevent the compounding of a contaminated product, one needs to have appropriate facilities, policies and procedures—and, importantly, the ability to train, assess, and maintain competency for employees on aseptic technique. Facilities and policies are readily assessed, but procedures to ensure that pharmacists and technicians—every employee—always follow strict aseptic technique on every preparation is difficult. The usual method for training, assessing, and maintaining competency is for the employee to be observed while making various preparations. Over time, the employee gets signed off as competent and is then allowed to prepare compounded sterile products with little to no oversight. Observation has been demonstrated to be a poor technique for assuring accuracy in compounding.11 Each employee must also complete annual media fill tests, simulating different compounded preparations.12 These tests require making different products using tryptic soy broth as a medium; if the employee demonstrates poor technique, the medium becomes contaminated. While the concept seems to be sound, research demonstrates that it is difficult to seed a positive. This research involved evaluating media fill tests of sensitivity and specificity when using poor technique.13 A total of 250 simulated compounded preparations were prepared. The first manipulation (25 preparations) followed best-practice aseptic technique and sterile compounding procedures. Each of the following 4 sets of manipulations removed one aspect of best-practice aseptic technique to the point whereby the preparation was made without any alcohol to sterilize the vial septum, the preparer used no gloves, the preparation was made outside of the compounding hood, and the uncapped vial was left 24 hours in ‘dirty’ air before preparation. Even though such poor conditions were utilized, no single preparation showed signs of turbidity, sedimentation, or visible microbial growth. A 0% contamination rate was documented. If the only option for ensuring that an employee is competent is through observatio