Systems and methods for automated profile identification
A system including at least an imaging device and a computing device configured to control the at least an imaging device while pre-scanning the target, receive, from the at least an imaging device, the pre-scan imaging data, detect a position of the fiducial marker within the pre-scan imaging data, input the pre-scan imaging data into a feature-learning neural network, output at least an image feature from the feature-learning neural network as a function of the pre-scan imaging data, wherein the at least a feature represents at least a specimen type, selecting a target profile as a function of the specimen type and the fiducial marker, wherein the target profile includes one or more imaging parameters and adjust, using the target profile, one or more imaging device parameters as a function of the one or more imaging parameters.
1 . A system for automated target profile identification, wherein the system comprises:
at least an imaging device configured to pre-scan a target to generate pre-scan imaging data, wherein the target comprises a fiducial marker; and
a computing device communicatively connected to the at least an imaging device,
wherein the computing device comprises:
at least a processor; and
a memory communicatively connected to the at least a processor, wherein the memory contains instructions configuring the at least a processor to:
control the at least an imaging device while pre-scanning the target;
receive, from the at least an imaging device, the pre-scan imaging data;
detect a position of the fiducial marker within the pre-scan imaging data;
input the pre-scan imaging data into a feature-learning neural network;
output at least an image feature from the feature-learning neural network, as a function of the pre-scan imaging data, wherein the at least a feature represents at least a specimen type;
select a target profile as a function of the specimen type and the fiducial marker, wherein the target profile comprises one or more imaging parameters; and
adjust, using the target profile, an imaging device configuration as a function of the one or more imaging parameters;
capture, using the at least an imaging device, imaging data as a function of the one or more imaging parameters determined from the pre-scan imaging data;
receive, using the at least a processor, high magnification imaging feedback from the at least an imaging device; and
verify and correct, using the at least a processor, a predicted target profile
as a function of the high magnification imaging feedback.
2 . The system of claim 1 , wherein receiving, from the at least an imaging device, the pre-scan imaging data comprises segmenting the pre-scan imaging data using a segmentation algorithm.
3 . The system of claim 1 , wherein the at least an imaging device comprises a plurality of imaging devices configured to simultaneously capture imaging data of a plurality of targets.
4 . The system of claim 1 , wherein:
the target profile comprises an imaging mode comprising one or more imaging parameters selected from a group of resolution, exposure time, illumination intensity, and scanning speed; and
the imaging mode is selected as a function of the specimen type to optimize image acquisition of the target.
5 . The system of claim 1 , wherein the target profile comprises a focus strategy, wherein the focus strategy comprises:
determining an optimal focal plane as a function of the target; and
adjusting focus parameters as a function of the optimal focal plane.
6 . The system of claim 1 , wherein verifying and correcting a predicted target profile as a function of the high magnification imaging feedback comprises:
generating refined profile classification data using the high magnification imaging feedback and a classification model;
comparing the refined profile classification data and the predicted target profile; and
updating the predicted target profile and corresponding one or more imaging parameters as a function of comparing the refined profile classification data and the predicted target profile.
7 . The system of claim 1 ,
wherein the at least a processor is further configured to classify, using a classification model, the at least a feature into a target type; and
wherein, the classification model is trained on a dataset comprised of images of target types, wherein the target types comprise one or more of microbiology samples, hematology samples, cytology samples, and histopathology samples.
8 . The system of claim 1 , wherein the at least a processor is further configured to extract data from the target using optical character recognition, wherein:
the target comprises a slide label; and
the extracted data is used to select the target profile.
9 . A method for automated target profile identification, wherein the method comprises:
controlling the at least an imaging device while pre-scanning a target;
receiving, from the at least an imaging device, pre-scan imaging data comprising a fiducial marker;
detecting a position of the fiducial marker within the pre-scan imaging data;
inputting the pre-scan imaging data into a feature-learning neural network;
outputting at least an image feature from the feature-learning neural network, as a function of the pre-scan imaging data, wherein the at least a feature represents at least a specimen type;
selecting a target profile as a function of the specimen type and the fiducial marker, wherein the target profile comprises one or more imaging parameters; and
adjusting, using the target profile, an imaging device configuration as a function of the one or more imaging parameters; and
capturing, using the at least an imaging device, imaging data, as a function of the one or more imaging parameters determined from the pre-scan imaging data;
receiving, by the at least a processor, high magnification imaging feedback from the at least an imaging device; and
verifying and correcting a predicted target profile as a function of the high magnification imaging feedback.
10 . The method of claim 9 , wherein receiving, from the at least an imaging device, the pre-scan imaging data comprises segmenting the pre-scan imaging data using a segmentation algorithm.
11 . The method of claim 9 , wherein the at least an imaging device comprises a plurality of imaging devices configured to simultaneously capture imaging data of a plurality of targets.
12 . The method of claim 9 , wherein:
the target profile comprises an imaging mode comprising one or more imaging parameters selected from a group of resolution, exposure time, illumination intensity, and scanning speed; and
the imaging mode is selected as a function of the specimen type to optimize image acquisition of the target.
13 . The method of claim 9 , wherein the target profile comprises a focus strategy, wherein the focus strategy comprises:
determining an optimal focal plane as a function of the target; and
adjusting focus parameters as a function of the optimal focal plane.
14 . The method of claim 9 , wherein verifying and correcting a predicted target profile as a function of the high magnification imaging feedback comprises:
generating refined profile classification data using the high magnification imaging feedback and a classification model;
comparing the refined profile classification data and the predicted target profile; and
updating the predicted target profile and corresponding one or more imaging parameters as a function of comparing the refined profile classification data and the predicted target profile.
15 . The method of claim 9 , further comprising classifying the at least a feature into a target type using a classification model, wherein the classification model is trained on a dataset comprised of images of target types, wherein the target types comprise one or more of microbiology samples, hematology samples, cytology samples, and histopathology samples.
16 . The method of claim 9 , further comprising extracting data from the target using optical character recognition, wherein:
the target comprises a slide label; and
the extracted data is used to select the target profile.
17 . A system for automated target profile identification, wherein the system comprises:
at least an imaging device configured to pre-scan a target to generate pre-scan imaging data, wherein the target comprises a fiducial marker; and
a computing device communicatively connected to the at least an imaging device, wherein the computing device comprises:
at least a processor; and
a memory communicatively connected to the at least a processor, wherein the memory contains instructions configuring the at least a processor to:
control the at least an imaging device while pre-scanning the target;
receive, from the at least an imaging device, the pre-scan imaging data;
detect a position of the fiducial marker within the pre-scan imaging data;
input the pre-scan imaging data into a feature-learning neural network;
output at least an image feature from the feature-learning neural network, as a function of the pre-scan imaging data, wherein the at least a feature represents at least a specimen type;
select a target profile as a function of the specimen type and the fiducial marker, wherein the target profile comprises one or more imaging parameters; and
adjust, using the target profile, an imaging device configuration as a function of the one or more imaging parameters;
perform inline validation, wherein performing inline validation comprises:
periodically capturing images of reference targets; and
comparing the images of reference targets against pre-defined imaging standards to verify and maintain imaging accuracy.
18 . A method for automated target profile identification, wherein the method comprises:
controlling the at least an imaging device while pre-scanning a target;
receiving, from the at least an imaging device, pre-scan imaging data comprising a fiducial marker;
detecting a position of the fiducial marker within the pre-scan imaging data;
inputting the pre-scan imaging data into a feature-learning neural network;
outputting at least an image feature from the feature-learning neural network, as a function of the pre-scan imaging data, wherein the at least a feature represents at least a specimen type;
selecting a target profile as a function of the specimen type and the fiducial marker, wherein the target profile comprises one or more imaging parameters; and
adjusting, using the target profile, an imaging device configuration as a function of the one or more imaging parameters; and
performing inline validation, wherein performing inline validation comprises:
periodically capturing images of reference targets; and
comparing the images of reference targets against pre-defined imaging standards to verify and maintain imaging accuracy.