System and method associated with determining physician attribution related to in-patient care using prediction-based analysis

A computer-implemented method associated with determining physician attribution related to in-patient care based at least on prediction of attribution values associated with patient-physician attribution, the method comprising:

• executing, on a processing device, the following operations:
• receiving extracted clinical data associated with a patient from a first database, the extracted clinical data comprising clinical information related to the patient, the clinical information including one or more sets of physician notes for the patient and a number of physicians assigned to the patient, the physician notes being indicative of clinical progress of the patient at various stages of treatment;
• assigning a vector value to the one or more sets of physician notes associated with the patient and assigning a feature value to the number of physicians assigned to the patient;
• determining a weight vector value which represents a level of priority associated with each individual physician note of the one or more sets of physician notes;
• determining initial attribution values using the attribution manager, the initial attribution values being based on a predictive analysis using predetermined weights associated with the vector values and the feature value;
• generating a training data set using a machine learning algorithm, the training data set being based on learned weight values associated with the clinical progress of the patient, the learning algorithm iterating through each data set and adjusting the weight values to maximize precision thereof;
• iteratively updating the vector values of the patient at least using a predictive analysis associated with the training data set using the adjusted weight values;
• eliminating erroneous predictions that are detected during the predictive analysis associated with the training data set;
• determining updated weight values when the erroneous predictions are detected and eliminated during the predictive analysis in order to improve prediction accuracy;
• processing the updated vector values of the patient and the updated weight values in order to increase precision and accuracy of the clinical information associated with the patient;
• determining patient-physician attribution values based on the updated vector values and the updated weight values that increases precision of the training data set thereby cross-validating the clinical information based on elimination of erroneous predictions and improved accuracy of the predictive analysis associated with the training data set;
• generating a clinical representation on a user interface device that includes validated patient-physician attribution values related to the in-patient care, the clinical representation including detailed attribution values associated with the patient, an indication of validated attribution values, and resultant attribution scores of each individual physician assigned to the patient;
• and
• providing remote access to users over a network, through the user interface device, so that any one of the users can access data relating to the clinical representation associated with the patient.

A non-transitory computer readable device storing instructions that, when executed by a processing device, cause the processing device to perform operations that include:

• receiving extracted clinical data associated with a patient from a first database, the extracted clinical data comprising clinical information related to the patient, the clinical information including one or more sets of physician notes for the patient and a number of physicians assigned to the patient, the physician notes being indicative of clinical progress of the patient at various stages of treatment;
• assigning a vector value to the one or more sets of physician notes associated with the patient and assigning a feature value to the number of physicians assigned to the patient;
• determining a weight vector value which represents a level of priority associated with each individual physician note of the one or more sets of physician notes;
• determining initial attribution values using an attribution manager, the initial attribution values being based on a predictive analysis using predetermined weights associated with the vector values and the feature value;
• generating a training data set using a machine learning algorithm, the training data set based on learned weight values associated with the clinical progress of the patient, the learning algorithm iterating through each data set and adjusting the weight values to maximize precision thereof;
• iteratively updating the vector values of the patient at least using a predictive analysis associated with the training data set using the adjusted weight values;
• eliminating erroneous predictions that are detected during the predictive analysis associated with the training data set;
• determining updated weight values when the erroneous predictions are detected and eliminated during the predictive analysis in order to improve prediction accuracy;
• processing the updated vector values of the patient and the updated weight values in order to increase precision and accuracy of the clinical information associated with the patient;
• determining patient-physician attribution values based on the updated vector values and the updated weight values that increases precision of the training data set thereby cross-validating the clinical information based on elimination of erroneous predictions and improved accuracy of the predictive analysis associated with the training data set;
• generating a clinical representation on a user interface device that includes validated patient-physician attribution values related to the in-patient care, the clinical representation including detailed attribution values associated with the patient, an indication of validated attribution values, and resultant attribution scores of each individual physician assigned to the patient;
• and
• providing remote access to users over a network, through the user interface device, so that any one of the users can access data relating to the clinical representation associated with the patient.