Artificial Intelligence System for Recognition of Occupational Lung Disease

Details

• Personal Author:
  Bailey K ; Coffman P ; Harber P ; McCoy JM ; Shimozaki S
• Description:
  An artificial intelligence occupational history system (AIOHS), a microcomputer based expert system, was developed to facilitate the recognition of occupational lung disease (OLD). The knowledge base (KB) structure was defined to capture relevant data using only two KB components: Objects (Job, Industry, Agent, Disease) and relations between pairs of primary objects of like or dissimilar type). Unlike traditional categorization schemes (e.g., SIC, SOC, ICDA), this permits direct use of the natural language of worker-patients. Forward chain search from case-specific data develops a network of possible causal associations, and backtracking then determines likelihood of causation using predicate calculus methods and disease class models (time, latency, dose-response relation, type, and duration). Additional data (e.g., respirator use) are requested by AIOHS only when relevant. A weighted binary search front-end interface uses SOC codes to find possible matches for jobs. (Empiric testing by two non-professional users with 100 jobs reported by asthmatics showed a mean user rating of accuracy of machine coding of 18.7 using a scale ranging from 1 = exact, through 20 = synonym, to 90 = uncertain). The AIOHS is a C-based expert system, combining features of slot-and-frame, graph, network, forward chain, backward chain, and predicate calculus systems. As a heuristic, it helps define the process used by OLD specialists, and as a practical tool it may facilitate the suspicion of OLD by non-occupational health specialists. [Description provided by NIOSH]
• Subjects:
  Asthma Asthma, Occupational Diagnosis Lung Diseases Occupational Health Respiratory System Respiratory Tract Diseases Safety Software
• Keywords:
  Artificial Intelligence Computer Applications Diagnostic Techniques Lung Disease Occupational Medicine Occupational Respiratory Disease
• ISSN:
  0003-0805
• Document Type:
  Text
• Funding:
  Grant
• Genre:
  Abstract or Summary
• Place as Subject:
  California ; OSHA Region 9
• CIO:
  National Institute for Occupational Safety and Health (NIOSH)
• Topic:
  Workplace Safety & Health
• Location:
  North America
• Volume:
  139
• Issue:
  4
• NIOSHTIC Number:
  nn:20058583
• Citation:
  Am Rev Respir Dis 1989 Apr; 139(4):A552
• Federal Fiscal Year:
  1989
• Performing Organization:
  University of California Los Angeles, Los Angeles, California
• Peer Reviewed:
  False
• Start Date:
  19870501
• Source Full Name:
  American Review of Respiratory Disease
• End Date:
  19930630
• Collection(s):
  National Institute for Occupational Safety and Health
• Main Document Checksum:
  urn:sha-512:0749d378e5b5ab7458d0a07dc07bafaa72e403dd4fcb62de6538c9840a4507c134bce5ca247642892c4e9158aa0294a201b035dd83bdd229e16d2ce56a9d0378
• Download URL:
  https://stacks.cdc.gov/view/cdc/221642/cdc_221642_DS1.pdf
• File Type:
  [PDF - 96.43 KB ]