Systems and methods for diagnosing a stroke condition

A system for quantitatively estimating a likelihood of a stroke condition of a subject, the system comprising:

• a database, containing classified sampled datasets acquired from a plurality of subjects, each positively diagnosed with at least one stroke condition, defining a positive stroke dataset;
• and
• a processor, configured to receive non-invasive clinical measurement data pertaining to said subject, and acquired from at least one sensor that is configured to acquire at least one of image data, sound data, movement data, and tactile data pertaining to said subject, said processor configured to construct, via machine learning in an initial training phase, a positive stroke model from at least part of said positive stroke dataset and, in a steady-state operation phase, continuously update by training through machine learning said positive stroke model via its defining parameters through parameter estimation and optimization, to extract from said clinical measurement data, potential stroke features according to at least one predetermined stroke assessment criterion;
• to compare said potential stroke features with said classified sampled datasets;
• and to determine according to said positive stroke model, without neuroimaging of said subject, a probability of a type of said stroke condition, and a probability of a corresponding stroke location of said stroke condition with respect to a particular brain location of said subject.

A system for quantitatively estimating a likelihood of a stroke condition of a subject, the system comprising:

• a client device including:
• at least one sensor, configured to acquire at least one of image data, sound data, movement data, and tactile data, all of which constitute non-invasive clinical measurement data pertaining to said subject;
• a user interface, configured to provide an indication of a probability for a type of said stroke condition, and a probability of a corresponding stroke location of said stroke condition with respect to a particular brain location of said subject, without neuroimaging of said subject;
• and
• a communication module, enabled for communication with a remote computer, said communication module configured to send said clinical measurement data to said remote computer, and to receive from said remote computer said indication;
• wherein said indication is based on a comparison between potential stroke features extracted from said clinical measurement data according to at least one predetermined stroke assessment criterion, with classified sampled data in a database acquired from a plurality of subjects, each positively diagnosed with at least one stroke condition, defining a positive stroke dataset, and based on a positive stroke model constructed via machine learning in an initial phase from at least part of said positive stroke dataset that is in a steady-state operation phase continuously updated by training machine learning via its defining parameters through parameter estimation and optimization.