Information processing apparatus, information processing method, and information processing program

1 . An information processing apparatus comprising:

a memory, storing designation information which designates, for each record item, a plurality of trained models set as a plurality of derivation methods and a preset priority order corresponding to each of the plurality of derivation methods, and exchange data which indicates a correspondence relationship between an item name actually used in a first medical document data and an item name used at a time of training of the plurality of trained models; and

at least one processor,

wherein the processor is configured to:

acquire the first medical document data including patient information related to a patient,

acquire the designation information stored in the memory to derive record information to be recorded in the record item related to the patient;

specify, by using the exchange data, the patient information of the item name actually used in the first medical document data that corresponds to the item name used at the time of training for a first trained model; wherein the patient information for the first trained model includes a first combination of items;

derive the record information for the record item by inputting the first combination of items into the first trained model set as a first derivation method, which is selected according to the preset priority order from among the plurality of derivation methods designated through the designation information;

in a case where at least one item in the first combination is omitted and the record information is not capable of being derived by using the first trained model:

specify, by using the exchange data, the patient information of the item name actually used in the first medical document data that corresponds to the item name used at the time of training for a second trained model, wherein the second trained model is set as a second derivation method having a lower priority in the preset priority order with respect to the first derivation method, and the patient information for the second trained model includes a second combination of items;

derive the record information for the record item by inputting the second combination of items into the second trained model,

wherein the plurality of trained models are trained in advance for performing determination or classification according to missing information of the patient information by using a plurality of combinations of the patient information and the record information, which are obtained for a plurality of patients, as training data, wherein a number and combination of items of the patient information to be input are different from each other among the plurality of the trained models;

generate second medical document data in which the derived record information is recorded in the record item; and

output a second medical document corresponding to the second medical document data from a printer in response to a request of a user.

2 . The information processing apparatus according to claim 1 ,

wherein the plurality of derivation methods for deriving the record information further comprise: diversion from the patient information, and performing determination or classification based on a predetermined rule regarding the patient information.

3 . The information processing apparatus according to claim 1 ,

wherein the processor is configured to present the medical document data by associating information, which indicates the derivation method applied to derive the record information from among the plurality of derivation methods, with the record information.

4 . The information processing apparatus according to claim 1 ,

wherein the processor is configured to present the medical document data by associating the patient information, which is used to derive the record information, with the record information.

5 . The information processing apparatus according to claim 1 ,

wherein the processor is configured to receive a designation input of the designation information.

6 . The information processing apparatus according to claim 1 ,

wherein the medical document data is data obtained by converting documents, which are related to hospital admission and discharge of the patient, into data.

7 . An information processing method executed by at least one processor included in an information processing apparatus, the method comprising:

storing designation information which designates, for each record item, a plurality of trained models set as a plurality of derivation methods and a preset priority order corresponding to each of the plurality of derivation methods in a memory; and exchange data which indicates a correspondence relationship between an item name actually used in a first medical document data and an item name used at a time of training of the plurality of trained models;

acquiring the first medical document data including patient information related to a patient;

acquiring the designation information stored in the memory to derive record information to be recorded in the record item related to the patient;

specifying, by using the exchange data, the patient information of the item name actually used in the first medical document data that corresponds to the item name used at the time of training for a first trained model; wherein the patient information for the first trained model includes a first combination of items;

deriving the record information for the record item by inputting the first combination of items into the first trained model set as a first derivation method, which is selected according to the preset priority order from among the plurality of derivation methods designated through the designation information;

in a case where at least one item in the first combination is omitted and the record information is not capable of being derived by using the first trained model:

specifying, by using the exchange data, the patient information of the item name actually used in the first medical document data that corresponds to the item name used at the time of training for a second trained model, wherein the second trained model is set as a second derivation method having a lower priority in the preset priority order with respect to the first derivation method, and the patient information for the second trained model includes a second combination of items;

deriving the record information for the record item by inputting the second combination of items into the second trained model,

wherein the plurality of trained models are trained in advance for performing determination or classification according to missing information of the patient information by using a plurality of combinations of the patient information and the record information, which are obtained for a plurality of patients, as training data, wherein a number and combination of items of the patient information to be input are different from each other among the plurality of the trained models;

generating second medical document data in which the derived record information is recorded in the record item; and

outputting a second medical document corresponding to the second medical document data from a printer in response to a request of a user.

8 . A non-transitory computer-readable storage medium storing an information processing program causing at least one processor included in an information processing apparatus to execute a process comprising:

storing designation information which designates, for each record item, a plurality of trained models set as a plurality of derivation methods and a preset priority order corresponding to each of the plurality of derivation methods in a memory; and exchange data which indicates a correspondence relationship between an item name actually used in a first medical document data and an item name used at a time of training of the plurality of trained models;

acquiring the first medical document data including patient information related to a patient;

acquiring the designation information stored in the memory to derive record information to be recorded in the record item related to the patient;

specifying, by using the exchange data, the patient information of the item name actually used in the first medical document data that corresponds to the item name used at the time of training for a first trained model; wherein the patient information for the first trained model includes a first combination of items;

deriving the record information for the record item by inputting the first combination of items into the first trained model set as a first derivation method, which is selected according to the preset priority order from among the plurality of derivation methods designated through the designation information;

in a case where at least one item in the first combination is omitted and the record information is not capable of being derived by using the first trained model:

specifying, by using the exchange data, the patient information of the item name actually used in the first medical document data that corresponds to the item name used at the time of training for a second trained model, wherein the second trained model is set as a second derivation method having a lower priority in the preset priority order with respect to the first derivation method, and the patient information for the second trained model includes a second combination of items;

deriving the record information for the record item by inputting the second combination of items into the second trained model,

wherein the plurality of trained models are trained in advance for performing determination or classification according to missing information of the patient information by using a plurality of combinations of the patient information and the record information, which are obtained for a plurality of patients, as training data, wherein a number and combination of items of the patient information to be input are different from each other among the plurality of the trained models;

generating second medical document data in which the derived record information is recorded in the record item; and

outputting a second medical document corresponding to the second medical document data from a printer in response to a request of a user.