Techniques for image-based monitoring of blood glucose status

An apparatus, comprising:

• at least one memory;
• and
• logic coupled to the at least one memory, the logic to:
• receive, as an input, quantitative patient monitoring information associated with a diabetic treatment of a patient, the patient monitoring information comprising at least one of a blood glucose level information, insulin dosage information, or insulin-on-board (IOB) information,
• generate a graph from the patient monitoring information, the graph comprising a temporal axis identifying historical readings of the patient monitoring information and a dependent axis showing the patient monitoring information for the historical readings,
• generate an image of at least a portion of the graph, the image comprising at least one digital image of the graph at a region of interest,
• process the image of the at least a portion of the graph using a computational model to determine, as an output, a predicted blood glucose condition of the patient, the blood glucose condition comprising one or more of a hypoglycemic episode, a hyperglycemic episode, a low blood sugar episode, a high blood sugar episode, or a normal blood sugar episode, the computational model configured based on historical monitoring information graph images to identify features representing an imminent blood glucose condition, and the processing comprising applying a filter to the image of the graph at the region of interest to determine if a trend in the graph at the region of interest matches the features, and
• administer a medicament with the apparatus based on the determined blood glucose condition.