Diagnosing rare diseases is like solving a puzzle, with physicians trying to fit together many complex clues.
Unfortunately, the 10% of the population who suffer from a rare disease wait an average of over seven years to find answers to their symptoms. With over 7,000 rare diseases to consider, matching a diagnosis to a patient's symptoms is a daunting task. By using new technologies to streamline the capture and analysis of phenotypic data in the diagnostic process, physicians are beginning to reduce the time it takes to identify a patient's disease while improving outcomes.
More and more hospitals with a rare disease focus are moving to more objective and advanced ways to evaluate unknown cases. The historic process of diagnosing a patient involved a physician sifting through the many phenotypes of the patient, such as diabetes and short stature, and using their own intuition while searching the overwhelming sea of publications to discover plausible diagnoses.
In the era of cognitive computing and vision science, new technologies can capture and analyze patient phenotypes in a more objective way while providing instant insights into patient genetics and differential diagnoses.
The rare disease burden is heavy, with one in ten people having a rare disease. Phenotypic information is a vital component to diagnosing these patients. Eighty percent of rare diseases have a genetic cause, and every rare disease has associated phenotypes. Technologies that capture phenotypes in a structured way allow for the use of those phenotypes by clinicians and other systems during the diagnostic process. Bioinformatics systems can readily use structured phenotypes to narrow down the genetic causes of the patient's symptoms. There is growing knowledge on associations between phenotypes and rare diseases that can support improved clinical decisions as well.
Systems that don't allow for structured capture and storage of phenotypic and genomic information lose the advantages. When a patient exhibits phenotypes that may indicate they are suffering from a rare disease, such as neuropathy or short stature, clinicians and labs must begin gathering and reviewing a massive amount of information in the quest to explain the symptoms. This analysis is the most time-consuming and costliest part of the process, resulting in the commonly quoted phrase: "A $1,000 genome with a $10,000 analysis."
Unfortunately, most EMRs are not designed to support the specific details of genetics and phenotyping in a structured way, so new tools have begun to emerge to fill the gap. Phenotypes can be captured in a structured way using standardized ontology like Human Phenotype Ontology (HPO), which has been broadly adopted for use in genomic diagnostics. HPO has also been mapped to related syndromes and genes within London Medical Databases, providing a source to correlate phenotypes, syndromes and genes. The Face2Gene suite of phenotyping apps has integrated the capture and storage of phenotypes, including facial analysis technology, analyzing the data to highlight clinically relevant syndromes and genes.
Rather than spending months manually reviewing unstructured phenotypic data, hospitals are now integrating technologies and new approaches to automate the capture and analysis of these phenotypes in an objective way. With integration to existing electronic medical record (EMR) systems, this newly captured phenotype and genotype information helps bring all the pieces together until the diagnostic puzzle is finally solved.
###
The views, opinions and positions expressed within these guest posts are those of the author alone and do not represent those of Becker's Hospital Review/Becker's Healthcare. The accuracy, completeness and validity of any statements made within this article are not guaranteed. We accept no liability for any errors, omissions or representations. The copyright of this content belongs to the author and any liability with regards to infringement of intellectual property rights remains with them.