Scientific evidence in court should be appraised for its validity before being presented to juries, the chairman of a Commons committee has said.
A new report of the Science and Technology Committee raised concerns about miscarriages of justice based on expert testimony.
Chairman Dr Ian Gibson highlighted the example of paediatrician Sir Roy Meadow, who figured in a number of cases involving baby deaths.
Sir Roy created the "Meadow's law" on cot deaths "that one in a family is a tragedy, two is suspicious and three is murder".
The theory formed part of the case against Angela Cannings, who was convicted of killing her two sons.
This was quashed last year leading to the review of hundreds of other cases.
Yesterday Dr Gibson said Sir Roy's claims were described as "nonsense" by statistical experts, but their evidence was not allowed.
"We want to find a better way to make sure juries and non-juries and certainly judges and lawyers are informed about the evidence," he said.
He added: "One thing would be to meet beforehand to see how good the evidence is, so that the opposite sides of the argument were presented and if there was a consensus available that would be fine.
"Lawyers admit that they don't understand it anyway so it becomes an issue of whether somebody with a knighthood is ranged up against say a post-doctoral student."
"In that case, it's all about charisma, experience, handling themselves in court. It shouldn't be, in terms of hard- nosed evidence."
The committee's report, Forensic Science on Trial, said miscarriages of justice could be prevented by "simple improvements" to the criminal justice system.
The MPs said the number of miscarriages of justice associated with problems of expert evidence is low and they were concerned about the lack of safeguards to prevent this happening.
They called for a Scientific Review Committee to be set up and a Forum for Science and Law to allow scrutiny of expert evidence.
This should also improve communication between the scientific and legal communities, they said.
The committee said it was "entirely unsatisfactory" that there was no agreed test for validating scientific evidence before being presented to court.
It called for jury research to be allowed to provide an insight of how juries handled scientific evidence.
The MPs suggested that trials without juries should be considered in cases which rely on "highly complex and technical forensic evidence".