Self-trainable neural network application for anomaly detection of biopharmaceutical products
System for analyzing anomalies in pharmaceuticals includes a server configured to host a neural network having an inference engine and a training engine, a database of images of in-process biologics; a first user interface module for displaying to a user particle morphologies in the images; a second user interface module for displaying to the user a training of the neural network; a third user interface module for displaying to the user an inference of images chosen by the neural network to fit selected criteria, wherein the neural network is a convolutional neural network, and training includes providing test images to the training engine to teach the neural network to recognize specific particle morphologies. The user provides images of the in-process biologics from the database, and the inference engine identifies anomalous particle morphologies in the user-provided images. A fourth user interface module provides a report about particle morphologies in the images.
1 . A method for analyzing anomalies in pharmaceuticals, the method comprising:
pre-processing one or more images that include pharmaceutical samples;
receiving, from an end user, designations of regions of interest in the one or more images;
automatically generating, by an active contour model, contours within the regions of interest designated by the user;
receiving, from the end user, labels identifying the contoured regions of interest;
training a machine learning model using the contoured regions of interest automatically generated by the active contour model and the corresponding labels identified by the user; and
deploying the trained machine learning model to identify an anomaly in a new pharmaceutical sample image.
2 . The method of claim 1 , further comprising:
receiving, from the end user, one or more fine tuning parameters for fitting the generated contours and the corresponding regions of interest.
3 . The method of claim 1 , further comprising:
saving the trained machine learning model with average data point parameters associated with the end user designated regions of interest.
4 . The method of claim 3 , wherein the deployment of the trained machine learning model comprises applying the average data point parameters to the new pharmaceutical sample image.
5 . The method of claim 3 , wherein the average data point parameters are based on corresponding areas of the end user designated regions of interest.
6 . The method of claim 1 , further comprising:
outputting the identified anomaly in the new pharmaceutical image.
7 . The method of claim 1 , further comprising:
providing a first interface for the end user to upload the one or more images that include pharmaceutical samples.
8 . The method of claim 7 , further comprising:
providing a second interface for the end user to designate the regions of interest in the one or more images.
9 . The method of claim 8 , further comprising:
providing a third interface for the end user to enter the labels identifying the contoured regions of interest.
10 . A method for analyzing anomalies in pharmaceuticals, the method comprising:
pre-processing, by an artificial intelligence model, one or more images that include pharmaceutical samples, autonomously generating regions of interest;
receiving, from an end user, labels identifying the autonomously generated regions of interest;
training a machine learning model using the autonomously generated regions of interest by the artificial intelligence model and the corresponding labels identified by the user; and
deploying the trained machine learning model to identify an anomaly in a new pharmaceutical sample image.
11 . The method of claim 10 , wherein the autonomous generation of the regions of interest comprises:
using an unsupervised model on the one or more images to autonomously generate the regions of interest.
12 . The method of claim 10 , wherein training the machine learning model comprises:
training the machine learning model using contours associated with the regions of interest, wherein the contours are provided as additional channel input.
13 . The method of claim 10 , further comprising:
outputting the identified anomaly in the new pharmaceutical image.
14 . The method of claim 10 , further comprising:
providing a first interface for the end user to upload the one or more images that include pharmaceutical samples.
15 . The method of claim 10 , further comprising:
providing a second interface for the end user to enter the labels identifying the autonomously generated regions of interest.
16 . A system for analyzing anomalies in pharmaceuticals, the system comprising:
a non-transitory computer readable medium storing computer program instructions;
at least one processor configured to execute the computer program instructions that when executed by the at least one processor cause the system to perform operations comprising:
autonomously generating, by an artificial intelligence model, regions of interest in one or more images that include pharmaceutical samples;
receiving, from an end user, labels identifying the generated regions of interest;
training a machine learning model using the autonomously generated regions of interest by the artificial intelligence model and the corresponding labels identified by the user; and
deploying the trained machine learning model to identify an anomaly in a new pharmaceutical sample image.
17 . The system of claim 16 , wherein generating regions of interest in the one or more images that include pharmaceutical samples comprises:
using an unsupervised model on the one or more images to autonomously generate the regions of interest.
18 . The system of claim 16 , wherein generating regions of interest in the one or more images that include pharmaceutical samples comprises:
receiving, from an end user, designations of regions of interest in the one or more images.
19 . The system of claim 16 , wherein the operations further comprise:
outputting the identified anomaly in the new pharmaceutical image.
20 . The system of claim 16 , wherein the operations further comprise:
providing an interface for the end user to upload the one or more images that include pharmaceutical samples.