For more than 20 years, James Duncan has insisted he was wrongfully convicted of aggravated child abuse for breaking his infant son's bones in 1993.
Now a Florida judge will decide if there's enough new medical evidence to give him a new trial.
Crime, law enforcement and corrections
Crimes against persons
Diseases and disorders
Health and medical
Law and legal system
Trial and procedure
Musculoskeletal disorders and injuries
Families and children
Family members and relatives
Health and health care (by demographic group)
Maternal and child health
Medical fields and specialties
Population and demographics
Business and industry sectors
Business, economy and trade
Health care professionals
Physicians and surgeons
Wounds and injuries
Thursday, Judge Michael Andrews held an evidentiary hearing in Pinellas County, Florida, to evaluate Duncan's case and what his team said is newly discovered evidence that could have resulted in an acquittal at his original trial in 1996.
Duncan, who also goes by Jim, and his wife took their infant son, Kody, to the emergency room in 1993, and doctors discovered he had 13 broken bones and a skull fracture.
Duncan was convicted of child abuse and sentenced to 70 years in prison. He's maintained his innocence in the years since.
Duncan's attorneys are relying on a 2014 study that could suggest Kody's injuries were the result of a metabolic bone disease like rickets, and not the result of child abuse.
"The basis of Mr. Duncan's new evidence claim is this article," defense attorney Lisabeth Fryer said Thursday.
But expert witnesses called by the state cast doubt on the peer-reviewed study, and said its findings were not widely accepted.
The first witness was Duncan's son and the alleged victim, Kody Duncan, who is now in his 20s and coaches tennis at a college in Pennsylvania. He believes his father should go free.
Since the conviction, father and son have seen each other only at one hearing more than a decade ago. Kody has never been allowed to visit his father in prison because he is allegedly the victim of his father's abuse, but they have phone calls once a week.
"I believe my dad is innocent," Kody told the court, before he was cut off by an objection from the prosecution.
Duncan smiled at his son as he left the stand.
Defense: Baby could have had fragile bones
Dr. David Ayoub, one of the authors of the study on which Duncan and his attorneys have based their case, testified for the defense as an expert witness.
According to the article, a type of fracture called a "classic metaphyseal legion," known as CML, that is commonly associated with child abuse has "features similar to infantile rickets."
Ayoub, a diagnostic radiologist, has said he believes Kody had a case of infantile rickets, a disease in which bones do not mineralize properly.
That could mean Kody's bones were very fragile, and that a simple act, like dressing a child or holding them for a vaccine, could result in broken bones.
According to Ayoub, there were shortcomings with earlier studies that identified CML fractures as high indicators of child abuse and argued the injury is actually a sign of a metabolic bone disease, like rickets.
Ayoub acknowledged that these fractures did not guarantee the child had not also been abused, but said these CML breaks were not a sign of it.
"Kody could have had healing infantile rickets, or copper deficiency," Ayoub testified.
"There is no question there is metabolic bone disease," he said. "There is no question."
The defense also called Dr. Marvin Miller, a pediatrician and medical geneticist. He was one of Ayoub's co-authors on the peer-reviewed study, which he said was the first to challenge the widely accepted medical theory that CMLs were "unequivocally" caused by child abuse.
Miller, testifying for the defense by video, told the court he studied Kody's medical records and found the injuries were "highly unlikely to be child abuse."
He pointed to the fact that Kody's medical reports at the time didn't identify any bruising or underlying tissue damage as a sign that the injuries were caused by problems with his bones.
Miller also said that based on psychological profiles done of Duncan, and the fact he had another son who wasn't abused, "My gut feeling is that he's a low risk to commit child abuse."
Theory not widely accepted, experts say
Ayoub and Miller's position is not widely accepted, witnesses called by the state testified.
Mainstream science does not recognize healing rickets as a legitimate cause for CMLs, which are typically associated with child abuse.
In a rebuttal to Ayoub's article, which was not presented in Thursday's hearing, the Child Abuse Committee of the Society for Pediatric Radiology said bone breaks of that type are "highly specific for child abuse."
"To deny this fact is to disregard the extensive experience and research of generations of pediatric radiologists," they wrote.
Expert witnesses called by the state agreed that Ayoub and Miller's study fell outside the mainstream.
"There is a lot of scientific evidence that refutes their positions and opinions," said Dr. Sally Smith, a pediatrician and medical director for the Pinellas County Child Protection Team.
She said Kody's fractures were the result of child abuse, and that most cases of abuse in children Kody's age do not have bruises, and tissue damage doesn't always happen.
Smith said it was possible for babies to get rickets, but it was rare in the United States, because babies get calcium and vitamin D from a variety of nutritious sources.
And if a baby does get rickets, it takes "many months, usually at least six months, before there's any overt signs" of rickets, Smith said. That's several months older than Kody was at the time of his injuries.
Prosecutors also called Dr. Mark Morris, a pediatrician who evaluated Kody in 1993 at All Children's Hospital, and testified against Duncan at his trial.
It was clear to him that Kody did not suffer from a disease like rickets.
"There are specific things you look for in metabolic bone disease that were not seen in this case," he said.
Kody had normal blood chemistry and X-rays, Morris said.
"A lot of different things were thrown out as possible, but there weren't any recognized chemical or radiological signs found in this child."
Morris also told the court that Ayoub's research was not widely accepted.
"There is a small group of people who agree with that, but the widespread opinion is this is not correct analysis of that injury," he told the court.
'I am encouraged'
Now the judge will have to decide whether the evidence Duncan presented could have resulted in an acquittal if it had been available in 1996. If so, the conviction will be overturned and a new trial ordered.
Duncan's family is cautiously optimistic.
"It was mentally draining, physically draining, too, but I am encouraged. I'm going to keep faith in God and hopefully things will go well," Kody told CNN after the hearing.
Duncan did not speak in court, but earlier in a jailhouse interview told CNN he was innocent and did not harm his son.
The judge told lawyers he would provide an update on his progress October 30.
On Saturday, Duncan called CNN from the Pinellas County Jail, where he's being held until the judge makes a decision on the case, which could take weeks. Duncan is nervous, he said, but hopeful.
He was grateful that so many of his family and friends came from around the country to attend Thursday's hearing, packing the moderately sized courtroom.
But he was most struck by his son Kody, whom he hasn't seen since he was a little boy, Duncan said.
"He's such a good looking young man," Duncan said. "I just wanted to hug him."