Medical data processing apparatus, medical data processing method, and non-transitory computer medium storing computer program

1. A medical data processing apparatus, comprising: processing circuitry configured to calculate a probability that received target image data has a lesion and a confidence interval indicating a reliability of the calculated probability; and output data on the lesion based on the calculated probability and the confidence interval of the probability, wherein the processing circuitry is further configured to acquire statistical data on a frequency of lesions aggregated based on characteristics of a plurality of image data, calculate the probability that the target image has the lesion and the confidence interval indicating the reliability of the probability, based on an arbitrary range including a measured value of the target image data using a frequency distribution indicated by the acquired statistical data, and output the data on the lesion based on the calculated probability and the confidence interval of the probability.

2. The medical data processing apparatus according to claim 1 , wherein the processing circuitry is further configured to calculate a range of a feature to which a feature of the target image data belongs as a counting range.

3. The medical data processing apparatus according to claim 2 , wherein the processing circuitry is further configured to: acquire a number of cases belonging to the counting range for each group in the plurality of the image data, and calculate an average value of probabilities of belonging and the a confidence interval for at least one group.

4. The medical data processing apparatus according to claim 1 , wherein the processing circuitry is further configured to output, as the data on the lesion, a predicted result regarding the lesion of the target image data based on a relationship between the calculated probability and the confidence interval of the calculated probability.

5. The medical data processing apparatus according to claim 4 , wherein the processing circuitry is further configured to output a determination result according to a prediction condition of the confidence interval.

6. The medical data processing apparatus according to claim 1 , wherein the processing circuitry is further configured to output, as the data on the lesion, the calculated probability, and the confidence interval of the calculated probability.

7. The medical data processing apparatus according to claim 1 , wherein the processing circuitry is further configured to collect the statistical data on the frequency of the lesion based on the characteristics of the plurality of image data.

8. The medical data processing apparatus according to claim 1 , wherein the processing circuitry is further configured to classify the received target image data into a plurality of image patterns, and calculate a confidence interval indicating the lesion probability of each image pattern as a classification reasonability scale, extract a part of the image patterns from which the lesion can be easily identified from the plurality of image patterns based on the classification reasonability scale of each image pattern of the plurality of image patterns, and preferentially display a part of the image patterns on a display.

9. The medical data processing apparatus according to claim 8 , wherein the processing circuitry is further configured to input the received target image data into a trained model that generates the classification reasonability scale based on the received target image data, thereby generating the classification reasonability scale.

10. The medical data processing apparatus according to claim 8 , wherein the processing circuitry is further configured to set a region of interest (ROI) of the target image data, and classify the target image data into a plurality of image patterns based on the ROI.

11. The medical data processing apparatus according to claim 8 , wherein the processing circuitry is further configured to display a name of a part of the image patterns as character data.

12. The medical data processing apparatus according to claim 8 , wherein the processing circuitry is further configured to use a predicted benign rate or a width of the confidence interval indicating a predicted malignancy rate as the classification reasonability scale.

13. The medical data processing apparatus according to claim 8 , wherein the processing circuitry is further configured to: predefine a specific image pattern to be displayed as a criterion regardless of a priority, display the specific image pattern on the display, and preferentially display the part of the image patterns on the display.

14. The medical data processing apparatus according to claim 8 , wherein the processing circuitry is configured to the probability of having the lesion is a benign probability or a malignant probability of mass.

15. A medical data processing method comprising: calculating a probability that received target image data has a lesion and a confidence interval indicating a reliability of the calculated probability, and outputting data on the lesion based on the calculated probability and the confidence interval of the probability, wherein the method is further comprises: acquire statistical data on a frequency of lesions aggregated based on characteristics of a plurality of image data, calculating the probability that the target image has the lesion and the confidence interval indicating the reliability of the probability, based on an arbitrary range including a measured value of the target image data using a frequency distribution indicated by the acquired statistical data, and outputting the data on the lesion based on the calculated probability and the confidence interval of the probability.

16. A non-transitory computer readable medium storing a program that when executed by circuitry, causes the circuitry to perform a method comprising: calculating a probability that received target image data has a lesion and a confidence interval indicating a reliability of the calculated probability, and outputting data on the lesion based on the calculated probability and the confidence interval of the probability, wherein the method further comprises to acquiring statistical data on a frequency of lesions aggregated based on characteristics of a plurality of image data, calculating the probability that the target image has the lesion and the confidence interval indicating the reliability of the probability, based on an arbitrary range including a measured value of the target image data using a frequency distribution indicated by the acquired statistical data, and outputting the data on the lesion based on the calculated probability and the confidence interval of the probability.