'I fucking changed science in one moment'

Riddled with 17 gunshot wounds, an 8-year-old child hit the operating table on what appeared to be the brink of death. She had no pulse in one leg, blood was pooling in her chest, and her abdomen was torn open.

The OR was in a state of chaos. Trauma surgeons had only moments to assess the child’s vitals and what course of action to take. Seventeen bullet wounds meant seventeen potential sources of death—and time was running out.

This is where “Abby,” a dietitian, statistician, and physiologist, stepped in. According to an Instagram post on her account (@the.fitness_scientist), Abby built a real-time statistical model of the patient’s vitals and survival odds—a series of calculated, science-driven moves that ultimately saved the young girl’s life.

Vitals became variables

When the patient destabilized before being operated on, Abby ran decision trees to determine what the most probable cause of sudden heart failure would be.

She knew that each calculation—each second—could mean the difference between life and death.

The model said stay abdominal

Using her statistical model, Abby ranked the patient’s gunshot wounds according to their likelihood of causing death. This ultimately guided surgeons on which wound to operate on first.

When the operating team was confronted with quickly needing to determine whether to open the patient’s chest or control her abdominal bleeding, Abby rushed to feed new data into her model.

The femoral artery was the best bet

Factoring in the patient’s vitals and the entry points of her wounds, combined with population-level trauma data, Abby’s model assessed which vessel in the patient’s body had most likely been hit.

The model pointed to the femoral artery with 83% probability, indicating that the operating team should focus on this part of the patient’s body.

Surgeons clamped the artery, and the patient’s blood circulation returned to her leg.

Modeling transfusion thresholds

When the patient went into metabolic acidosis, Abby’s model factored in the patient’s base deficit and body weight to calculate the specific hemoglobin levels at which a blood transfusion would be recommended.

By estimating the volume of fluid the patient had lost and the number of units of blood she needed, Abby’s model communicated to the surgical team how—and when—to act.

What saved the patient? 

Reflecting on the role her statistical modeling played in the young patient's survival, Abby wrote in her Instagram post, “I wasn’t in the OR to observe. I was there to direct decisions. Every number, every second … I translated chaos into action. I did the math that helped keep her alive.”

This, Abby says, is how the use of science in the medical field can save lives: “This is what science looks like: Not a headline. Not a theory. But equations solving chaos. Data driving scalpels. Models saving children.”