Method and apparatus for classifying detection inputs in medical images

A method of classifying detection inputs as belonging to a first class or a second class depending on a numerical score associated with each detection input, wherein the detection input comprises a digital image, comprising the steps of:

• segmenting each detection input into a multiplicity of regions of interest;
• scoring the multiplicity of regions of interest to produce a set of scores;
• setting a first threshold that is the score of a known true positive input in a training set used to train an algorithm for scoring the regions of interest;
• comparing each of the scores with the first threshold,
• if at least one score is less than the first threshold, classifying in the first class all regions of interest having a score that is less than a second threshold, wherein the second threshold has a value greater than the first threshold, and
• if no score is less than the first threshold, classifying each region of interest in the second class.

A method of classifying detection inputs as belonging to a first class or a second class depending on a numerical score associated with each detection input, wherein the detection input comprises a digital image, comprising the steps of:

• segmenting each detection input into a multiplicity of regions of interest;
• scoring the multiplicity of regions of interest to produce a set of scores;
• comparing each of the scores with a first threshold,
• if at least one score exceeds the first threshold, lowering a second threshold that is lower than the first threshold, or if no score exceeds the first threshold, raising the second threshold, and
• classifying in the first class all regions of interest having a score that exceeds the second threshold.

A method of classifying detection inputs as belonging to a first class or a second class depending on a numerical score associated with each detection input, wherein the detection input comprises a digital image, comprising the steps of:

• segmenting each detection input into a multiplicity of regions of interest;
• scoring the multiplicity of regions of interest to produce a set of scores;
• comparing each of the scores with a first threshold,
• if at least one score is less than the first threshold, raising a second threshold that is greater than the first threshold, or if no score is less than the first threshold, lowering the second threshold, and
• classifying in the first class all regions of interest having a score that are less than a second threshold.

A method of identifying regions of interest in a medical image to detect a medical condition of a subject, the medical condition being characterized by prominent instances of the medical condition within the subject and minor instances of the medical condition within the subject, the method comprising:

• segmenting the medical image into a multiplicity of regions;
• scoring the multiplicity of regions to produce a set of scores for each region;
• defining a first threshold value indicating detection of a prominent instance of the medical condition;
• comparing each of the scores of the multiplicity of regions with the first threshold;
• if at least one score exceeds the first threshold value, classifying in a first class all regions of interest having a score that exceeds a second threshold value, wherein the second threshold value has a value lower than the first threshold value;
• and
• if no score exceeds the first threshold value, classifying each region of interest in a second class.

A method of detecting embolisms in a subject by identifying regions of interest in a computed tomography image of the subject, the embolisms being characterized by prominent embolisms and minor embolisms, the method comprising:

• segmenting the computed tomography image into a multiplicity of regions;
• scoring the multiplicity of regions to produce a set of scores for each region;
• defining a first threshold value indicating detection of a prominent embolism;
• defining a second threshold value indicating detection of a minor embolism;
• comparing each of the scores of the multiplicity of regions with the first threshold;
• if at least one score exceeds the first threshold value, classifying in a first class all regions of interest having a score that exceeds the second threshold value, wherein the second threshold value has a value lower than the first threshold value, and
• if no score exceeds the first threshold value, classifying each region of interest in a second class.