SimulConsult receives SBIR I from the National Institute of Arthritis and Musculoskeletal and Skin Diseases of the NIH beginning September 2013 with Michael M Segal MD PhD as Primary Investigator
This project responds to a critical shortage of pediatric rheumatologists by enlisting top rheumatologists to encapsulate the diagnostic information in their field in the most advanced decision support software tool for diagnosis. The project tests workability of integration into a tool used in other areas of medicine and assesses the benefits of such rheumatology assistance for both rheumatologists and general clinicians. By doing so, it tests whether this decision support approach can be used more widely to improve accuracy and cost effectiveness in medicine.
Published in Genome Web on May 18, 2012 by Uduak Grace Thomas
SimulConsult, a bioinformatics company based in Chestnut Hill, Mass., is marketing a new software tool called the Genome-Phenome Analyzer, which lets clinicians and research labs combine whole-genome and whole-exome sequencing data with clinical findings and phenotype information
Published in Neurology on May 15, 2012 by Segal MM and Schiffmann R.
The editorial discusses the role for diagnostic decision support in facilitating an evidence-based discussion between clinicians and payers.
Published in “Pediatric Neurology: Principles and Practice”, Swaiman KF et al., editors, 5th edition, Saunders, Chapter 108 by Segal M and Lever S
Published in MIT’s Technology Review on September 21, 2011.
The article discusses IBM’s Watson, Isabel and SimulConsult in the context of medical decision support.
SimulConsult receives contract from the National Library of Medicine of the NIH beginning September 27, 2010 with Michael M. Segal MD PhD as Primary Investigator
The project is designed to assess the feasibility of a computational approach to diagnostic decision support based on an open, comprehensive database encompassing many diseases, using a Bayesian model with an evidence-based authoring process in which it is possible to update one disease in isolation, without commenting on other diseases, and do so with detailed temporally-sensitive information.
Published in Brain on July 12, 2010 by Peter Garrard John Stephenson Vijeya Ganesan Timothy Peters
The authors used SimulConsult to analyze a historical diagnosis.
Published as a book in September 2009 by King MD and Stephenson BP, Mac Keith Press
In the book the authors discuss SimulConsult, beginning on page 1 with “With the explosion of electronic information such as found on PubMed and the availability of freely accessible diagnostic software such as SimulConsult one might ask what possible use is there for a handbook like this.
In an interview with Hugh Hewitt, Clayton Christensen comments on how SimulConsult will disrupt aspects of the health system.
Published in the Health Business Group Blog on March 6, 2009.
In The Computer Will See You Now, Dr. Anne Armstrong-Coben highlights the depersonalization and mindless checkboxes of electronic health records. She identifies Tablet PCs as the way to deal with the depersonalization, but is silent on what to do about the mindless checkboxes:
Published on December 4, 2008 by Clayton Christensen et al
A book by Clayton Christensen and colleagues discusses SimulConsult as an example of a “disruptive innovation” that helps generalist and specialist doctors make diagnoses at the level of specialists and sub-specialists.
Published in Acta Pediatric in November 12, 2007 by Michael M Segal, MD PhD
Textbooks teach about diagnosis by describing diseases and the collection of findings in each disease. However, patients show up with findings, not diseases, and we must invert our knowledge to take the patient’s collection of findings and figure out the correct diagnosis. It is not so simple to do this, particularly in pediatrics. The Office of Rare Diseases at the National Institutes of Health estimates that 8% of people have one of the 7000 diseases listed as ‘rare’, and most of these diseases have onset in the pediatric years.
This is a lot of information to consider, and as a result diagnosis is wrong an estimated 15% of the time.
MD NetGuide article “Top Ten Physician Inventors” (on page 7) cites as #2 Michael Segal MD PhD: “Was responsible for the development of SimulConsult, a decision-support software that combines clinical and laboratory findings to help physicians come to a simultaneous diagnosis on a disease.”
Published in Neurology Today on February 6, 2007 by Orly Avitzur MD.
A sophisticated neurology wiki that helps doctors make clinical decisions is exactly what child neurologist Michael M. Segal, MD, PhD, had in mind when he created SimulConsult. SimulConsult users input data into software that helps them arrive at a differential diagnosis of neurological syndromes.
David J. Michelson, MD, assistant professor of neurology at Loma Linda University School of Medicine, said he has been amazed by SimulConsult’s ability to generate perfect matches to single conditions.
“The software is very user-friendly and intuitive,” he said. “It is designed to highlight metabolic and genetic syndromes which often only differ by certain specific findings.” … a 7-year-old boy with a neurodegenerative disorder with dementia, cerebellar atrophy, and spastic quadriplegia – despite EMG and muscle biopsy results suggesting denervation – was a perfect match for neuroaxonal dystrophy. He quipped, “Steven Ashwal made that diagnosis within five seconds of hearing me describe the child, which just proves that I need the program a lot more than he does.”
The tool has matured and increased in accuracy through a collaborative effort of many neurologists. In contrast to Wikipedia, all submissions to the SimulConsult database are reviewed by doctors following the existing norms for medical journals. Dr. Segal contends that wikis are more descriptive, providing details about findings in diseases, than prescriptive, providing rules tailored for particular clinical situations. The descriptive information available through wikis is a good way to aggregate evidence-based medicine, Dr. Segal added.
“Our software addresses 1,400 inherited and congenital neurological diseases, and uses over 24,000 data points,” said Dr. Segal. “Since there are no clinical rules, adding new material is straightforward and can be done merely by changing individual data points.”
SimulConsult also created a blog format for clinical cases in which medical terms from a case narrative are hyperlinked into the diagnostic software (simulconsult.com/cases/). Physicians can click into the software with all patient findings already entered. The cases blog format is being developed together with the Child Neurology Society for residents and other doctors in practice who want to keep their skills sharp and up-to-date with the latest advances in diagnostic knowledge.
The ability to click into information tools from narratives about patients has the potential for much wider application. “In essence, a medical chart is a privacy-protected blog,” Dr. Segal pointed out. If data such as labs, tests, history and physical exam findings can be filtered into software, physicians can be easily assisted in medical decision-making.”
“The cases blog illustrates the potential of a neurology wiki,” Dr. Segal continued. “As similar tools become an integral part of practice, opportunities for collecting the wisdom of the neurology community in these ways will increase.”
Published in Journal of Child Neurology in April 21, 2006 by Michael M Segal MD PhD
Medical practice will be improved by the use of software that not only assists with diagnosis but can do so at the bedside, where the doctor can act immediately upon suggestions such as useful findings to check. Medical education will shift away from a focus on details of unusual diseases and toward a focus on skills of physical examination and using compertized tools. Medical publishing, in contrast, will shift toward greater detail: it will be increasingly important to quantitate the frequency of findings in diseases and their time course since such information can have a major impact clinically when added to decision support software.
Neuronline review of SimulConsult by G. Fuller
Imagine the scene: you have just seen a patient with what seems likely to be an inherited or congenital syndrome who has a combination of symptoms and signs you do not recognise and in whom standard investigations have provided no clues. What do you do? Looking in the books is very time consuming and depends on how well the books are indexed; searching PubMed or online inheritance in man (OMIM) is often difficult using just symptoms and signs. Wouldn’t it be useful if you could put the clinical information into a database to narrow down the differential diagnosis? This is the idea that drives www.simulconsult.com, which describes itself as a “medical decision support software” that “allows doctors and other medical professionals to combine clinical and laboratory findings” and get a “simultaneous consult about diagnosis”. Registration is straightforward and this leads you to an introductory video and the opportunity to search the site. The searching is reasonably intuitive. You enter the age and sex, and then can add clinical feature and test results, either having searched for them or from an alphabetical or category based pick list.
by Orly Avitzur MD in the American Academy of Neurology’s publication “Neurology Today”:
James D. Reggin, MD, the only child neurologist at MeritCare Health System in Fargo, ND, was called into consultation for an infant delivered 11 days earlier by C-section, now in respiratory distress and intubated, when pediatricians observed persistent hypotonia and a poor sucking reflex. Dr. Reggin noted that the infant displayed dysmorphic features as well as severe microcephaly, syndactyly, and multiple capillary hemangiomas. When he could not pinpoint a specific syndrome, he turned to SimulConsult, a computerized decision-support tool that helps neurologists arrive at difficult diagnoses by entering a list of findings.
Despite having been in practice more than 12 years, he had never seen a case of Smith-Lemli-Opitz (SLO), the condition revealed to be “very high probability” by the program. Prompted by this discovery, he quickly obtained a serum cholesterol level and later confirmed the diagnosis of SLO by finding a marked elevation of 7-dehydroxycholesterol. Neurologists are increasingly using decision-support tools such as SimulConsult (simulconsult.com) to make diagnoses – and these and other resources improve patient safety.
American Academy of Neurology Neurology Today article “How Neurologists Use the Internet to Enhance Clinical Decision-Making” by Orly Avitzur MD. Excerpt:
If C. J. Malanga, MD, PhD had his way, neurologists would be required to – not left with the option of – learning how to use the Internet…. Dr. Malanga first discovered the benefits of the Internet as a child neurology fellow at Massachusetts General Hospital (MGH), trying to determine a differential diagnosis for a case posted on the weekly neuropathology conference. The case involved a 15-year-old boy who was first seen at age 5 with slurred speech and gait instability and who continued to deteriorate over the next 10 years with emerging marked dystonia, dysphagia, tics and seizures.
On their own, Dr. Malanga and fellow trainees came up with a long list of differential diagnoses. But after entering important variables in query form into a Web site, simulconsult.com, the search was narrowed considerably to Friedreich ataxia, Hallervorden-Spatz, and Huntington chorea (akinetic-rigid form). Autopsy revealed almost complete striatal atrophy with minimal cortical involvement. The final diagnosis, confirmed by CAG repeat count, was juvenile Huntington disease.
Note: Dr. Malanga ran this case through the software in May 2002. The case can be viewed in the current version of the software at this link and registered users can click into the software to see the differential diagnosis (only the clinical findings were used) and work with the case. Furthermore, adding the family history information (that the father was similarly affected but three siblings were not) using the family history feature added to the software after the Neurology Today article appeared, the correct diagnosis is elevated to #1 in the software by a wide margin.