System and method for correcting patient index
Inventors
Tse, Henry Tat Kwong • Shah, Ajay M. • Guillou, Lionel • Sheybani, Roya • Johnson, Audun • Landwehr, Peter • Stevenson, Hunter
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-11964281-B2
Publication Date
2024-04-23
Expiration Date
Abstract
A system and/or method for determining an immune activation state of a subject can include: deforming leukocytes within a microfluidic channel, acquiring a plurality of images of the leukocytes, determining biophysical parameters of the leukocyte, adjusting the biophysical parameters, and determining the immune activation state of the subject based on the biophysical parameters.
Core Innovation
The disclosed invention relates to determining a sepsis condition of a subject using a dataset generated from a blood sample in which leukocytes are deformed within a deformation region of a microfluidic channel. For each leukocyte, the dataset comprises a plurality of images of the leukocyte as it is deformed, and biophysical parameter(s) are determined for each leukocyte based on the plurality of images and associated per-leukocyte image features and trajectory parameters.
The dataset is filtered to exclude images comprising overlapping leukocytes. Overlapping leukocytes are detected based on a monotonicity in feature motion between images of the leukocyte, and additional filtering is described in terms of excluding based on entrance and exit positions in the deformation region using thresholding constraints, as well as excluding based on additional event/frame conditions such as isolation or occlusion and z-plane conditions.
The biophysical parameters are linearly adjusted based on at least one of a flow rate and a path traversed by the respective leukocyte. The adjustment can depend on leukocyte handedness, leukocyte cell type such as neutrophil, and z-plane, and it is used to compute a numerical index from the adjusted biophysical parameter dataset. Based on the numerical index, the method determines a probability that the subject is experiencing sepsis, including sepsis-related conditions, with example computation involving a logistic regression and comparison to a threshold.
Claims Coverage
The independent claim is directed to a method that determines a sepsis condition by analyzing deformation-region leukocyte image datasets, filtering overlap, deriving and linearly adjusting biophysical parameters, computing a numerical index, and determining a sepsis probability. The dependent claims refine the independent claim by adding specific filtering constraints, specifying adjustment dependencies, and constraining index/probability computation details.
Dataset reception from deformed leukocytes in a microfluidic channel
Receiving a dataset generated from a blood sample, where the dataset includes, for each leukocyte, a plurality of images as the leukocyte is deformed within a deformation region of a microfluidic channel.
Overlapping leukocyte exclusion based on monotonicity in feature motion
Filtering the dataset to exclude images comprising overlapping leukocytes, comprising detecting the overlapping leukocytes based on a monotonicity in feature motion between images of the leukocyte.
Biophysical parameter determination from plurality of per-leukocyte images
Determining a biophysical parameter dataset by, for each leukocyte, determining one or more biophysical parameter associated with the respective leukocyte based on the plurality of images associated with the respective leukocyte.
Linear adjustment of biophysical parameters based on flow rate and leukocyte path
Linearly adjusting the one or more biophysical parameters based on at least one of a flow rate and a path traversed by the respective leukocyte.
Numerical index computation and sepsis probability determination
Determining a numerical index based on the biophysical parameter dataset, and based on the numerical index, determining a probability that the subject is experiencing sepsis.
Entrance and exit position filtering in the deformation region
Filtering the dataset by determining entrance and exit positions of the leukocyte in the deformation region from multiple images and excluding images when the entrance/exit positions do not satisfy threshold distance constraints relative to expected entrance and expected exit positions.
Handedness-dependent linear adjustment
Adjusting the one or more biophysical parameters based on the leukocyte flow path such that they are linearly adjusted for a first handedness and not linearly adjusted for a second handedness.
Neutrophil z-plane dependent biophysical parameter adjustment
When the leukocyte cell type is a neutrophil, linearly adjusting the one or more biophysical parameters based on the z-plane of the leukocyte.
Numerical index based on neutrophil size and monocyte viscoelastic inertial response percentile
Determining a numerical index using an average size of neutrophils and a 75th percentile of a viscoelastic inertial response of monocytes.
Across the independent and dependent claims, the method operates on deformation-region leukocyte image datasets, removes overlapping leukocyte images via monotonicity in feature motion, derives per-leukocyte biophysical parameter(s), linearly adjusts those parameters based on flow rate and traversed path with additional dependencies, computes a numerical index, and then determines a probability of sepsis.
Stated Advantages
Documented Applications
No documented applications found
Interested in licensing this patent?