Testing a Computational Model to Estimate Patient Size
Conditions
Brief summary
A computational model has been created to estimate the abdominal depth of a patient from a single x-ray image. The model has been tested using phantoms and found to be accurate; this study aims to test the accuracy of the model with patients and in a clinical setting. This will be achieved by enrolling patient's who have already been referred for an anterior-posterior abdomen x-ray examination to the trial, taking a physical measurement of their anterior-posterior abdominal depth and then comparing this measured value with a value as estimated using the computational model based on the patient's x-ray image.
Detailed description
A non-commercial computational model has been developed in-house to estimate the patient's anterior-posterior or lateral depth using the radiographic image and the known exposure factors with which it was undertaken. This model has been tested using single composition phantoms and found to be accurate. If it was found to be accurate for real clinical examinations, this would automate the measurement of patient size and give institutions the estimate of patient size required for local paediatric patient dose audit. In turn, this would provide the national data required to propose national reference values for paediatric x-ray examinations, which would give all institutions an important comparator for their performance. This would lead to optimisation in those sites most requiring it; nationally, paediatric x-ray imaging would improve in time. This pilot study is necessary to determine if the computational model is accurate enough to be relied upon. Accuracy will be determined by comparing the estimate made by the computational model for each patient with an actual measurement of the patient's anterior-posterior abdomen depth made at the time of the examination.
Interventions
OTHERA single measurement of the patient's abdominal depth
A single measurement of the patient's anterior-posterior abdominal depth
Sponsors
University of Dundee
NHS Tayside
Study design
Observational model
CASE_ONLY
Time perspective
CROSS_SECTIONAL
Eligibility
Sex/Gender
ALL
Age
17 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* Adult * Referred to Ninewells Hospital for an anterior-posterior abdomen x-ray examination
Exclusion criteria
* Patients unable to give consent * Patients who have had a contrast injection in the previous 24 hours * Patients suffering abdominal pain at the time of the examination
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Accuracy of the Computational Model | 2 months | The computational model was used to estimate the patient's anterior-posterior abdominal depth using the digital radiographic image, the exposure factors with which it was acquired and a priori knowledge relating to the x-ray unit and digital detector. The outcome measure was the accuracy with which the computational model estimates the patient's anterior-posterior abdominal depth. It was determined by comparing the estimate to measured anterior-posterior abdominal depth (measured at the time of the x-ray examination). Results are expressed as a percentage deviation; a low % deviation is more accurate, a high % deviation less accurate. |
Countries
United Kingdom
Participant flow
Participants by arm
| Arm | Count |
|---|---|
| Patient Cohort 20 patients referred to Ninewells Hospital radiology department for an anterior-posterior abdomen x-ray examination. All of these patients will have a measurement of their anterior-posterior depth before undergoing their x-ray examination. An estimate of their anterior-posterior depth will then be made from their x-ray image using the computational model.
A single measurement of the patient's abdominal depth: A single measurement of the patient's anterior-posterior abdominal depth | 20 |
| Total | 20 |
Baseline characteristics
| Characteristic | Patient Cohort |
|---|---|
| Age, Categorical <=18 years | 0 Participants |
| Age, Categorical >=65 years | 5 Participants |
| Age, Categorical Between 18 and 65 years | 15 Participants |
| Ethnicity (NIH/OMB) Hispanic or Latino | 0 Participants |
| Ethnicity (NIH/OMB) Not Hispanic or Latino | 0 Participants |
| Ethnicity (NIH/OMB) Unknown or Not Reported | 20 Participants |
| Region of Enrollment United Kingdom | 20 participants |
| Sex: Female, Male Female | 10 Participants |
| Sex: Female, Male Male | 10 Participants |
Adverse events
| Event type | EG000 affected / at risk |
|---|---|
| deaths Total, all-cause mortality | 0 / 0 |
| other Total, other adverse events | 0 / 0 |
| serious Total, serious adverse events | 0 / 0 |
Outcome results
Primary
Accuracy of the Computational Model
The computational model was used to estimate the patient's anterior-posterior abdominal depth using the digital radiographic image, the exposure factors with which it was acquired and a priori knowledge relating to the x-ray unit and digital detector. The outcome measure was the accuracy with which the computational model estimates the patient's anterior-posterior abdominal depth. It was determined by comparing the estimate to measured anterior-posterior abdominal depth (measured at the time of the x-ray examination). Results are expressed as a percentage deviation; a low % deviation is more accurate, a high % deviation less accurate.
Time frame: 2 months
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Patient Cohort | Accuracy of the Computational Model | 5.8 percentage agreement (est/measured) | Standard Deviation 4.6 |