Systems and methods to generate a surgical risk score and uses thereof

A method for assessing immune activation of a patient for determining a surgical risk score indicative the patient's risk for a surgical complication following surgery, comprising:

• collecting one or more samples from the patient,
• obtaining an ex-vivo-stimulation sample, wherein the ex-vivo-stimulation sample is or is derived from the one or more samples from the patient, wherein the ex-vivo-stimulation sample comprises immune cells that are collected prior to a surgical procedure to be performed on the patient;
• measuring values of a plurality of omic biological features, wherein the plurality of omic biological features comprises a set of one or more immune-activation measurements of the patient, wherein each immune-activation measurement of the set is an amount of activation of an immune-related signaling protein yielded by ex vivo stimulation and is determined by:
• incubating the ex-vivo-stimulation sample with one or more immune activating agents selected from the group consisting of:
• a Toll-Like Receptor 4 (TLR4) agonist, interleukin-2 (IL-2), interleukin-4 (IL-4), interleukin-6 (IL-6), interleukin-1β (IL-1β), Tumor Necrosis Factor-α (TNF-α), Interferon-α (IFN-α), and phorbol 12-myristate 13-acetate (PMA)/ionomycin;
• and
• measuring an amount of activation of an immune-related signaling protein of the immune cells within the ex-vivo-stimulation sample by:
• contacting the immune cells with an affinity reagent that binds to the immune-related signaling protein in an activated state, and
• measuring an amount of binding of the affinity reagent to the immune-related signaling proteins in the activated state;
• wherein the measuring the amount of activation of the immune-related signaling protein is performed subsequent to incubating the ex-vivo-stimulation sample with one or more immune activating agents;
• and
• computing a surgical risk score for the patient by entering the values of the plurality of omic biological features into a machine learning model trained to generate surgical risk scores
• wherein the surgical risk score comprises a probability that the patient will develop a surgical site complication following the surgical procedure to be performed on the patient;
• wherein the machine learning model is trained using a bootstrap procedure on a plurality of individual data layers;
• wherein each data layer represents one type of data from the plurality of omic biological features and at least one artificial feature, wherein at least one data layer represents a set of one or more immune-activation measurements of the patient of which each immune-activation measurement of the set is the amount of activation of the immune-related signaling protein yielded by the ex vivo stimulation;
• wherein each data layer comprises data for a population of individuals;
• wherein each omic biological feature includes feature values for all individuals in the population of individuals;
• wherein, for a respective data layer, each artificial feature is a new variable obtained from a non-artificial feature among the plurality of omic biological features via a mathematical operation performed on the feature values of the non-artificial feature;
• wherein the mathematical operation being chosen among the group consisting of:
• a permutation, a sampling with replacement, a sampling without replacement,
• a combination, a knockoff and an inference
• wherein the machine learning model includes weights for a set of biological omic features of the plurality of omic biological features;
• wherein during training of the machine learning model and for each data layer, for every repetition of the bootstrap, initial weights are computed for the plurality of omic biological features and the at least one artificial feature associated with that data layer using an initial statistical learning technique, and at least one selected feature is determined for each data layer, based on a statistical criteria depending on the computed initial weights;
• and
• wherein the statistical criteria depends on significant weights among the computed initial weights;
• the significant weights being non-zero weights, when an initial statistical learning technique is a sparse regression technique;
• the significant weights being weights above a predefined weight threshold, when the initial statistical learning technique is a non-sparse regression technique
• wherein the statistical criteria is based on an occurrence frequency of the significant weights.