Methods and apparatus for evaluating developmental conditions and providing control over coverage and reliability

Abstract

The methods and apparatus disclosed herein can evaluate a subject for a developmental condition or conditions and provide improved sensitivity and specificity for categorical determinations indicating the presence or absence of the developmental condition by isolating hard-to-screen cases as inconclusive. The methods and apparatus disclosed herein can be configured to be tunable to control the tradeoff between coverage and reliability and to adapt to different application settings and can further be specialized to handle different population groups.

What is claimed is: 
     
         1 . A method for providing an evaluation of input data of an individual obtained through a computer system, said method comprising:
 (a) receiving said input data of said individual related to a disorder, delay, or impairment; and   (b) evaluating said input data using at least one machine learning model, thereby generating an evaluation result, wherein said evaluation result is a first categorical determination or a first inconclusive determination with respect to a presence or an absence of said disorder, delay, or impairment.   
     
     
         2 . The method of  claim 1 , wherein at least one threshold range for determining if an evaluation result is inconclusive is adjustable. 
     
     
         3 . The method of  claim 2 , wherein a threshold range of said first categorical determination decreases when a threshold range of said first inconclusive determination increases. 
     
     
         4 . The method of  claim 3 , wherein an accuracy of said first categorical determination increases when said threshold range of said first categorical determinations decreases. 
     
     
         5 . The method of  claim 2 , wherein said threshold range of said first categorical determination or said first inconclusive determination is based on an inclusion rate for said first categorical determination. 
     
     
         6 . The method of  claim 5 , wherein said first categorical determination for said presence or absence of said disorder, delay, or impairment in said individual is based on a specified sensitivity, a specified specificity, a specified negative predictive value, or a specified positive predictive value. 
     
     
         7 . The method of  claim 5 , wherein said inclusion rate is no less than 70%, and wherein said first categorical determination results in a sensitivity of at least 80% or a specificity of at least 80%. 
     
     
         8 . The method of  claim 1 , wherein said first categorical determination is selected from said presence or said absence of said disorder, delay, or impairment. 
     
     
         9 . The method of  claim 1 , wherein said at least one machine learning model comprises a subset of a plurality of tunable machine learning models. 
     
     
         10 . The method of  claim 9 , further comprising:
 (a) requesting additional data when said evaluation result comprises said first inconclusive determination; and   (b) generating a second categorical determination or a second inconclusive determination based on said additional data using at least one additional machine learning model selected from said plurality of tunable machine learning models.   
     
     
         11 . The method of  claim 9 , further comprising:
 (a) combining scores for each of said subset of said plurality of tunable machine learning models to generate a combined preliminary output score; and   (b) mapping said combined preliminary output score to said first categorical determination or to said first inconclusive determination for said presence or absence of said disorder, delay, or impairment in said individual.   
     
     
         12 . The method of  claim 11 , wherein said combined preliminary output score is based on a rule-based logic or a combinatorial technique for combining said scores. 
     
     
         13 . The method of  claim 1 , wherein said disorder, delay, or impairment comprises pervasive development disorder (PDD), autism spectrum disorder (ASD), social communication disorder, restricted repetitive behaviors, interests, and activities (RRBs), autism (“classical autism”), Asperger's Syndrome (“high functioning autism), PDD-not otherwise specified (PDD-NOS, “atypical autism”), attention deficit disorder (ADD), attention deficit and hyperactivity disorder (ADHD), speech and language delay, obsessive compulsive disorder (OCD), depression, schizophrenia, Alzheimer's disease, dementia, intellectual disability, or learning disability. 
     
     
         14 . The method of  claim 13 , wherein said disorder, delay, or impairment is autism spectrum disorder or autism. 
     
     
         15 . The method of  claim 1 , further comprising generating a personal therapeutic treatment plan for said individual based on said evaluation result. 
     
     
         16 . The method of  claim 15 , wherein said personal therapeutic treatment plan is generated using at least one statistical or machine learning model. 
     
     
         17 . The method of  claim 15 , further comprising receiving feedback data based on performance of said personal therapeutic treatment plan and updating said personal therapeutic treatment plan based on said feedback data. 
     
     
         18 . The method of  claim 17 , wherein said feedback data comprises at least one of efficacy, compliance, and response to said personal therapeutic treatment plan. 
     
     
         19 . The method of  claim 15 , wherein said personal therapeutic treatment plan comprises a drug therapy and a non-drug therapy. 
     
     
         20 . The method of  claim 19 , wherein said non-drug therapy comprises digital therapeutics.