Reduced interaction CT scanning
Systems and methods for a reduced interaction CT scanning workflow. A sensor is used to capture an image of a patient on the table. Scan parameters are automatically set. A full CT scan is performed without a scout scan. During the full CT scan, the scan parameters are adjusted based on the raw CT measurements from the full CT scan. A radiology report is automatically generated from the results of the full CT scan.
1. A method for reduced interaction medical imaging, the method comprising:
capturing, with a camera, surface data representing a patient;
registering, by a processor, the surface data to a model of the patient;
calculating, by the processor, one or more uncertainty measures based on the registration of the surface data to the model of the patient;
estimating, by the processor and a machine trained network configured to generate a generalized probability distribution of scan parameters, scan parameters for a CT scanner from the registered image data and the one or more uncertainty measures;
initiating a diagnostic scan by the CT scanner with the scan parameters without performing an additional scan;
adjusting, during the scan, the scan parameters as a function of acquired scan data during the scan and the one or more uncertainty measures; and
outputting a radiology report after completing the scan.
2. The method of claim 1 , wherein the scan parameters are estimated using a network trained using machine training.
3. The method of claim 1 , wherein the scan parameters comprise at least one of a start slice position, an end slice position, an expected absorption curve, or dose.
4. The method of claim 3 , wherein the start slice position and the end slice position correspond to anatomical landmarks identified on the registered surface data.
5. The method of claim 1 , wherein estimating the scan parameters comprises:
estimating, by the processor, a three-dimensional model of the patient; and
using, by the processor, markers on the three-dimensional model and a selected protocol to estimate the scan parameters.
6. The method of claim 1 , wherein the scan parameters are estimated by registering a prior CT scan of the patient to the surface data and using the prior CT scan to estimate the scan parameters.
7. The method of claim 1 , wherein the adjustment of scan parameters using machine training comprises comparing the predicted scan parameters with current measurements from the CT scanner as the scan is being generated and setting the scan parameters based on the comparison.
8. The method of claim 7 , wherein an end slice position of the scan parameters is adjusted as a function of current measurements indicating an unexpected change in density values in raw data from the CT scanner.
9. The method of claim 1 , further comprising:
stopping the scan when an estimate for the scan parameters with the one or more uncertainty measures are outside a range of errors.
10. The method of claim 1 , wherein the radiology report is generated using a machine trained network trained to process one or more regions of interest and classify any pathologies present.
11. A system for reduced interaction CT scanning without preforming an additional scan, the system comprising:
a sensor configured to capture surface data of a patient;
a CT scanner configured to scan the patient using one or more scan parameters and generate CT scan data; and
a processor configured to estimate, from the surface data of the patients and one or more uncertainty measures derived from the surface data, the one or more scan parameters using a machine trained network configured to generate a generalized probability distribution of scan parameters, initialize the CT scanner with the estimated one or more scan parameters, and adjust the one or more scan parameters during a scan of the patient by the CT scanner as a function of received CT scan data and the one or more uncertainty measures.
12. The system of claim 11 , wherein the processor is further configured to automatically generate a radiology report from an output of the CT scanner.
13. The system of claim 12 , wherein the system further comprises:
a display configured to display the radiology report.
14. A method for reduced interaction medical imaging, the method comprising:
capturing, with a three-dimensional camera, surface data representing an outer surface of a patient;
generating, by a processor, a synthetic topogram from the surface data, wherein the synthetic topogram includes uncertainty measures calculated as a function of a range of possible error values during generation;
generating, by dose modulation software, one or more scan parameters from the synthetic topogram without performing an additional scan using a machine trained network trained to obtain a generalized probability distribution of the scan parameters;
initiating a scan by a CT scanner with the one or more scan parameters;
adjusting, during the scan, the one or more scan parameters as a function of acquired scan data during the scan; and
outputting a radiology report after completing the scan.
15. The method of claim 14 , wherein adjusting the scan comprises: adjusting the scan parameters constrained by the range of error values.
16. The method of claim 14 , further comprising: registering, by the processor, the surface data to a patient model as a function of a prior scan; wherein the synthetic topogram is generated from the registered surface data.
17. The method of claim 14 , wherein outputting the radiology report comprises generating the radiology report using a machine trained model trained to classify one or more pathologies in a CT image.