Information processing device, information processing method, and computer program

Abstract

This information processing device comprises a model acquisition unit, a data of interest acquisition unit, and a determination execution unit. The model acquisition unit acquires a trained model in which medical record data before or during intervention of a predetermined treatment for patients with a specific disease is set as an explanatory variable and the necessity of palliative care for the patients is set as an objective variable. The data of interest acquisition unit acquires medical record data for a patient of interest. The determination execution unit: determines the necessity of palliative care for the patient of interest by inputting the medical record data for the patient of interest to the trained model; and causes an output device to output the result of the determination.

Description

Information Processing Apparatus, Information Processing Method, and Computer Program

[0001] The technology disclosed in this specification relates to information processing for determining the need for palliative care for patients with a specific disease.

[0002] Palliative care is holistic care aimed at improving the quality of life (QOL) of patients with severe diseases (such as cancer) and their families. With palliative care, improvements in QOL, reduction of symptom burden, and extension of survival period are expected. Conventionally, palliative care has often been provided as end-of-life care during the watchful waiting period just before death.

[0003] It has been proposed to interpret the output of a learned model that takes information at the time of admission as input and outputs information on whether a patient has died during hospitalization or their condition has deteriorated to the point where end-of-life care is required, for all inpatients, not limited to cancer patients, as the prediction result of the need for palliative care (see, for example, Non-Patent Document 1).

[0004] Dennis H Murphree, et al. 15 others, "Improving the delivery of palliative care through predictive modeling and healthcare informatics", Journal of the American Medical Informatics Association, American Medical Informatics Association, February 21, 2021, Vol. 28, No. 6, p. 1065 - 1073

[0005] For palliative care to be effective, it is recommended to provide palliative care to patients who need it from an early stage, such as before or during treatment intervention. However, determining the need for palliative care is difficult for non-specialists, and the number of palliative care specialists (palliative care specialists and certified palliative care nurses) is small. As a result, many patients who need palliative care from an early stage do not receive it because their need is not recognized. Furthermore, the conventional techniques mentioned above use information such as mortality rates and whether or not end-of-life care was required, and therefore cannot determine the need for early palliative care.

[0006] This specification discloses a technology capable of solving the above-mentioned problems.

[0007] The technologies disclosed herein can be implemented, for example, in the following forms:

[0008] (1) The information processing device disclosed herein comprises a model acquisition unit, a target data acquisition unit, and a determination execution unit. The model acquisition unit acquires a trained model in which medical record data before or during a predetermined treatment intervention for a patient with a specific disease is used as an explanatory variable and the need for palliative care for the patient is used as the objective variable. The target data acquisition unit acquires the medical record data for the target patient. The determination execution unit inputs the medical record data for the target patient into the trained model to determine the need for palliative care for the target patient and outputs the determination result to an output device.

[0009] This information processing device can use a trained model to accurately determine the need for palliative care from an early stage, such as before or during treatment intervention.

[0010] (2) In the above-mentioned information processing device, the medical record data may include an index value representing the degree of emotional distress. With this configuration, the need for early palliative care can be determined with even greater accuracy.

[0011] (3) In the above-mentioned information processing device, the medical record data may include an index value representing the degree of pain. With this configuration, the need for early palliative care can be determined with even greater accuracy.

[0012] (4) In the above-mentioned information processing device, the specific disease may be cancer. With this configuration, the need for early palliative care for cancer patients can be determined with high accuracy.

[0013] (5) In the above-mentioned information processing device, the predetermined treatment may be chemotherapy. With this configuration, the need for palliative care from an early stage, such as before or during chemotherapy intervention for cancer patients, can be determined with high accuracy.

[0014] (6) In the above-mentioned information processing device, if the determination execution unit determines in the determination result that there is a high need for palliative care for the target patient, it may display a screen on the display device, which is an output device, for inputting whether or not the patient wishes to receive palliative care. This configuration makes it possible to facilitate the provision of palliative care to patients who have a high need for palliative care.

[0015] (7) In the above-mentioned information processing device, the determination execution unit may determine the recommended content of palliative care to be provided to the target patient based on the degree of need for palliative care for the target patient in the determination result, and display the recommended content on the display device which is the output device. With this configuration, it is possible to provide patients with palliative care of appropriate content according to the degree of need for palliative care.

[0016] (8) Other information processing devices disclosed herein include a model acquisition unit. The model acquisition unit generates a trained model by performing training, in which medical record data before or during a predetermined treatment intervention for a patient with a specific disease is used as explanatory variables and the need for palliative care for the patient is used as the objective variable.

[0017] This information processing device can obtain a trained model that accurately determines the need for palliative care from an early stage, such as before or during treatment intervention.

[0018] Furthermore, the technologies disclosed herein can be implemented in various forms, for example, in the form of an information processing device, an information processing method, a computer program that implements such a method, or a non-temporary recording medium on which such a computer program is stored.

[0019] This embodiment includes an explanatory diagram illustrating the palliative care need determination model MO, an explanatory diagram illustrating the schematic configuration of the information processing device 100, a flowchart illustrating the palliative care need determination model acquisition process, a flowchart illustrating the labeling process, an explanatory diagram illustrating an example of items (explanatory variables) in the medical record data MD for learning data LD, a flowchart illustrating the determination process, an explanatory diagram illustrating the determination accuracy using the palliative care need determination model MO, and an explanatory diagram illustrating the determination accuracy using the palliative care need determination model MO.

[0020] (Embodiment) (Overview of the Palliative Care Needs Determination Model MO) Figure 1 is a schematic diagram illustrating the Palliative Care Needs Determination Model MO in this embodiment. The Palliative Care Needs Determination Model MO is a trained model for determining the need for specialized palliative care for patients with specific diseases. The following describes the case where the specific disease is cancer. The Palliative Care Needs Determination Model MO of this embodiment can also be applied to diseases other than cancer, such as interstitial pneumonia, chronic obstructive pulmonary disease (COPD), heart failure, renal failure, and neurological diseases (e.g., amyotrophic lateral sclerosis).

[0021] As shown in Figure 1, once a patient is diagnosed with cancer, a treatment method is selected, and treatment intervention is carried out according to the selected method. Examples of treatment methods include surgery, radiation therapy, and drug therapy (chemotherapy). If the cancer is cured, treatment is terminated, and the patient transitions to outpatient care for regular recurrence assessments. If a cure is not expected, treatment is also terminated by discontinuation. In the case of discontinuation of treatment, the patient transitions to the end-of-life care phase.

[0022] Traditionally, palliative care has often been provided as end-of-life care, during the final stages of life. However, for palliative care to be effective, it is recommended to provide it to patients who need it from an early stage, either before or during treatment intervention. Here, "early" refers to the period from diagnosis to the end of a predetermined period after the start of treatment, and is different from the end of life. The predetermined period may be, for example, two months, three months, or six months. After diagnosis and before treatment intervention, palliative care may be needed for physical symptoms (such as shortness of breath), psychological symptoms (such as insomnia), issues related to medical expenses, work, school, family, caregiving, and childcare. During treatment intervention, in addition to physical and psychological symptoms, palliative care may be needed for concerns about treatment choices, anxiety about treatment, and side effects of treatment.

[0023] The palliative care need determination model MO of this embodiment is a trained model that uses medical record data MD from before or during treatment intervention as an explanatory variable and the need for palliative care for the patient as an dependent variable. By using the palliative care need determination model MO of this embodiment, the need for palliative care can be determined at an early stage, such as before or during treatment intervention, and the provision of palliative care to patients who need it can be promoted. The output of the palliative care need determination model MO may represent the need for palliative care as a binary value (yes / no), as a multi-level score representing the degree of need for palliative care, or as a probability that palliative care will be needed.

[0024] (Configuration of Information Processing Device 100) Next, the configuration of the information processing device 100 for creating a palliative care need determination model MO and performing determination using the palliative care need determination model MO will be explained. Figure 2 is an explanatory diagram showing the schematic configuration of the information processing device 100. The information processing device 100 is composed of, for example, a computer (PC, server, smartphone, tablet terminal, etc.).

[0025] The information processing device 100 comprises a control unit 110, a storage unit 120, a display unit 130, an operation input unit 140, and an interface unit 150. Each of these units is connected to the others via a bus 190 so as to be able to communicate with each other.

[0026] The display unit 130 of the information processing device 100 is configured, for example, as a liquid crystal display, and displays various images and information. The display unit 130 is an example of an output device and a display device. The operation input unit 140 is configured, for example, as a keyboard, mouse, buttons, microphone, trackpad, etc., and receives operations and instructions from an administrator. The display unit 130 may also function as the operation input unit 140 by being equipped with a touch panel. The interface unit 150 is configured, for example, as a LAN interface or USB interface, and communicates with other devices by wired or wireless connection. The information processing device 100 may also be equipped with other output devices (for example, a speaker).

[0027] The storage unit 120 of the information processing device 100 is composed of, for example, ROM, RAM, a hard disk drive (HDD), and stores various programs and data, and is used as a working area and temporary storage area for data when executing various programs. For example, the storage unit 120 stores a judgment processing program CP, which is a computer program for executing various processes described later. The judgment processing program CP is provided, for example, stored on a computer-readable recording medium (not shown) such as a CD-ROM, DVD-ROM, or USB memory, or is provided in a state that can be obtained from an external device (a server on a network or other terminal device) via the interface unit 150, and is stored in the storage unit 120 in a state that can be operated on the information processing device 100.

[0028] The storage unit 120 of the information processing device 100 stores the learning data LD, the palliative care need determination model MO, and the determination result data RD either in advance or during the execution of various processes described later. The contents of this information and data will be explained in accordance with the descriptions of the various processes described later.

[0029] The control unit 110 of the information processing device 100 is configured, for example, with a CPU, and controls the operation of the information processing device 100 by executing a computer program read from the storage unit 120. For example, the control unit 110 functions as a determination processing unit 111 for executing various processes described later by reading and executing a determination processing program CP from the storage unit 120. The determination processing unit 111 includes a raw data acquisition unit 112, a training data acquisition unit 113, a model acquisition unit 114, a target data acquisition unit 115, and a determination execution unit 116. The functions of each of these units will be explained in accordance with the descriptions of various processes described later.

[0030] (Palliative Care Needs Determination Model Acquisition Process) Next, the palliative care needs determination model acquisition process performed by the information processing device 100 of this embodiment will be described. Figure 3 is a flowchart of the palliative care needs determination model acquisition process. The palliative care needs determination model acquisition process is the process of acquiring a palliative care needs determination model MO. In this embodiment, the information processing device 100 acquires the palliative care needs determination model MO by performing predetermined machine learning to create the palliative care needs determination model MO itself. The palliative care needs determination model acquisition process is started when a user operates the operation input unit 140 of the information processing device 100 and inputs a start command.

[0031] First, the raw data acquisition unit 112 (Figure 2) of the information processing device 100 acquires information used to create the palliative care need determination model MO (hereinafter referred to as "raw data Io") (S110). Raw data Io is the data that forms the basis of the learning data LD used for training, verification, and testing the palliative care need determination model MO. Raw data Io includes medical record data MD. Raw data Io includes, for example, in-hospital cancer registry data, medical expense billing data, nursing hospitalization record data, and in-hospital data warehouse data. Raw data Io is acquired via the interface unit 150 or via the operation input unit 140.

[0032] Next, the learning data acquisition unit 113 (Figure 2) of the information processing device 100 acquires learning data LD by performing preprocessing on the original data Io (S120). Examples of preprocessing include labeling, downsampling, and missing data imputation. The learning data LD is data that associates each patient with medical record data MD and whether or not they have a need for palliative care (for example, whether they have a need for palliative care "yes" or "no").

[0033] Figure 4 is a flowchart of the labeling process. First, a list of individuals to be evaluated is created (S210). For example, from the raw data Io, cancer patients aged 18 or older who have distant metastases or have been diagnosed with stage IV cancer and who have received chemotherapy and pain screening are extracted to create the list of individuals to be evaluated.

[0034] Next, for each patient included in the list of individuals to be evaluated, a classification is performed based on the results of the pain screening (S220), and a classification is performed based on the presence or absence of palliative care needs based on nursing records (S230, S250). Furthermore, if the pain screening result is positive and there is "no" palliative care need based on nursing records, a further classification is performed based on the presence or absence of palliative care needs based on expert evaluation (S240).

[0035] If the pain screening result is positive and there is a need for palliative care based on the nursing record, or if there is no need for palliative care based on the nursing record but there is a need for palliative care based on the expert assessment, the patient is labeled as having a need for palliative care (S260). If the pain screening result is positive and there is no need for palliative care based on the nursing record and there is no need for palliative care based on the expert assessment, the patient is labeled as having no need for palliative care (S270). On the other hand, if the pain screening result is negative and there is a need for palliative care based on the nursing record, the patient is labeled as having a need for palliative care (S260), and if there is no need for palliative care based on the nursing record, the patient is labeled as having no need for palliative care (S270).

[0036] Figure 5 is an explanatory diagram showing an example of the items (explanatory variables) in the medical record data MD in the training data LD. As shown in Figure 5, the medical record data MD includes, for example, the following main items: 1. Degree of emotional distress (NRS (Numerical Rating Scale) 0-10 points): Continuous variable 2. Degree of impairment in daily life due to emotional distress (NRS 0-10 points): Continuous variable 3. Degree of pain (NRS 0-10 points, 0-1 / 2-3 / 4-6 / 7-10 points): Categorical variable 4. ECOG performance scale (0-4 stages, 0 / 1 / 2 / 3-4): Categorical variable 5. Age: Continuous variable 6. Sex: Binary variable 7. Embryonic risk classification of chemotherapy regimen (0-3 points, 0 / 1 / 2 / 3): Categorical variable

[0037] Furthermore, as shown in Figure 5, the medical record data MD may include, for example, the following items: 8. Primary site of cancer (8 sites: head and neck, gastrointestinal tract, hepatobiliary pancreatic, gynecological, urinary tract, respiratory, skin, other): categorical variable (39 sites from ICD-10 codes and further consolidated into 8 classifications) 9. Order of anticancer drug treatment (first-line treatment where anticancer drug is administered for the first time: 1, second-line treatment where a different drug is used for cancer recurrence or progression: 2, etc.): continuous variable 10. First onset and recurrence of cancer: binary variable 11. Severity of physical symptoms other than pain (NRS 0-10 points, 0-1 / 2-3 / 4-6 / 7-10 points): categorical variable 12. Sodium level in blood sampling (3 divisions based on median and interquartile range, 0 / 1 / 2): categorical variable 13. Calcium level in blood sampling (3 divisions with equal distribution of people, 0 / 1 / 2): categorical variable 14. 15. LDH (lactate dehydrogenase) level in blood sample: continuous variable 16. Total protein level in blood sample (divided equally among the number of people into 3 groups, 0 / 1 / 2): categorical variable 17. Patient's key person (relationship classified into 6 categories: spouse, child, parent, sibling, in-law, friend / colleague, cohabitant): categorical variable 18. BMI: continuous variable 19. Presence or absence of peritoneal dissemination: binary variable 10. Means of transportation to the hospital (categorized into 4 categories: car, public transport, motorcycle, walking): categorical variable 21. Smoking index (number of cigarettes smoked per day × number of years smoking): continuous variable 22. Consultation preference (category 6 categories: no preference at present, palliative care team, certified oncology nurse, oncology counseling / social worker, discharge support, other): categorical variable 1. Department of medical care upon admission (13 categories: Breast, Respiratory Medicine / Surgery, Chemotherapy, Gynecology, Cardiology, Orthopedics, Urology / Nephrology, Gastroenterology, Gastrointestinal Surgery, Dermatology, Diabetes, General Medicine, Otolaryngology): Categorical variable 23. Number of days elapsed since diagnosis: Continuous variable 24. Reason for unplanned emergency medical admission (10 categories: Poor general condition, impaired consciousness, respiratory failure / heart failure, acute drug poisoning, shock, metabolic disorder, burns, trauma, emergency surgery, other): Categorical variable

[0038] Next, the model acquisition unit 114 (Figure 2) of the information processing device 100 creates a palliative care need determination model MO by performing machine learning using the training data LD (S130 in Figure 3). Various known machine learning algorithms can be used for the machine learning required to create the palliative care need determination model MO. The created palliative care need determination model MO is stored in the storage unit 120 of the information processing device 100. With this, the palliative care need determination model acquisition process (Figure 3) is completed.

[0039] (Decision Processing) Next, the decision processing performed by the information processing device 100 of this embodiment will be described. Figure 6 is a flowchart of the decision processing. The decision processing is a process that determines whether or not palliative care is necessary for the target patient using the palliative care need determination model MO. The decision processing may be performed by the same information processing device 100 in which the palliative care need determination model acquisition process described above is performed, or it may be performed by a different information processing device 100. The decision processing is started when a user operates the operation input unit 140 of the information processing device 100 and inputs a start command.

[0040] First, the target data acquisition unit 115 (Figure 2) of the information processing device 100 acquires medical record data MD for the target patient (S310). The medical record data MD for the target patient may be acquired, for example, from an in-hospital system, or it may be acquired based on input by the target patient or their family. Alternatively, the medical record data MD may be acquired via a telemedicine / diagnosis system.

[0041] Next, the determination execution unit 116 (Figure 2) of the information processing device 100 inputs the patient's medical record data MD into the palliative care need determination model MO to determine whether palliative care is necessary for the patient (S320). The determination result may be binary data representing whether palliative care is necessary, a score indicating the degree of need for palliative care, or the probability that palliative care will be needed. The determination execution unit 116 generates determination result data RD, which is information indicating the determination result, and stores it in the storage unit 120 of the information processing device 100.

[0042] Next, the determination execution unit 116 outputs a determination result based on the determination result data RD (S330). For example, the determination execution unit 116 causes the display unit 130 to display the determination result.

[0043] When it is determined in the determination result that the need for palliative care for the target patient is high, the determination execution unit 116 may cause the display unit 130 to display a screen for inputting the presence or absence of a desire for palliative care provision. Further, the determination execution unit 116 may determine the recommended content of palliative care to be provided to the target patient based on the high degree of need for palliative care for the target patient in the determination result, and cause the display unit 130 to display the determined recommended content.

[0044] For example, if the attending physician refers to the determination result displayed on the display unit 130 and the determination result is that palliative care is necessary (or the need for palliative care is high), the attending physician can explain palliative care to the target patient or introduce a palliative care team (for example, a group of specialists, certified nurses, certified pharmacists, physical therapists, occupational therapists, social workers, psychologists, dietitians, speech therapists, etc.). In addition, the patient and their family can refer to the determination result displayed on the display unit 130 and convey the presence or absence of a desire for palliative care to the attending physician or the like. Thus, the determination process is completed.

[0045] (Example) An example of the above-described palliative care need determination model MO will be described below.Figures 7 and 8 are explanatory diagrams showing the determination accuracy using the palliative care need determination model MO. In this example, a retrospective cohort study was conducted. The target patients were patients with advanced solid cancer aged 18 or older diagnosed with distant metastasis or stage IV, who received chemotherapy during the period from April 1, 2018 to March 31, 2023 and underwent pain screening. The number of sample patients n = 561. 80% of the samples were used as training data to create the palliative care need determination model MO, and 20% were used as test data to verify the accuracy of the palliative care need determination model MO. XGBoost was used as the machine learning algorithm for creating the palliative care need determination model MO. The output of the palliative care need determination model MO was binary data indicating the presence or absence of the need for palliative care.

[0046] As shown in FIGS. 7 and 8, in the verification result of the palliative care need determination model MO using test data, the sensitivity was 1.00, the specificity was 0.81, and the ROC-AUC was 0.90, and it was possible to determine the necessity of palliative care with high accuracy.

[0047] (Effect of this embodiment) As described above, the information processing apparatus 100 of this embodiment includes a model acquisition unit 114, a target data acquisition unit 115, and a determination execution unit 116. The model acquisition unit 114 acquires a palliative care need determination model MO, which is a learned model using medical record data MD before or during a predetermined treatment intervention for a patient with a specific disease as an explanatory variable and the necessity of palliative care for the patient as an objective variable. The target data acquisition unit 115 acquires medical record data MD for the target patient. The determination execution unit 116 inputs the medical record data MD for the target patient into the palliative care need determination model MO to determine the necessity of palliative care for the target patient and outputs the determination result to the display unit 130.

[0048] According to this embodiment, it is possible to accurately determine the necessity of palliative care from an early stage such as before or during treatment intervention using the palliative care need determination model MO.

[0049] In this embodiment, the medical record data MD may include an index value representing the degree of the patient's mental distress. With this configuration, it is possible to more accurately determine the necessity of palliative care from an early stage.

[0050] In this embodiment, the medical record data MD may include an index value representing the degree of pain. With this configuration, it is possible to more accurately determine the necessity of palliative care from an early stage.

[0051] In this embodiment, the specific disease may be cancer. With this configuration, it is possible to accurately determine the necessity of palliative care from an early stage for cancer patients.

[0052] In this embodiment, the prescribed treatment may be chemotherapy. This configuration allows for highly accurate determination of the need for palliative care from an early stage, such as before or during chemotherapy intervention for cancer patients.

[0053] In this embodiment, if the determination execution unit 116 determines in the determination result that the patient has a high need for palliative care, it may display a screen on the display unit 130 for inputting whether or not the patient wishes to receive palliative care. This configuration makes it possible to facilitate the provision of palliative care to patients who have a high need for it.

[0054] In this embodiment, the determination execution unit 116 may determine the recommended content of palliative care to be provided to the target patient based on the degree of need for palliative care in the determination result, and display the recommended content on the display unit 130. With this configuration, the patient can be provided with palliative care of appropriate content according to the degree of need for palliative care.

[0055] (Examples of application) The following are examples of applications of the technology disclosed herein.

[0056] (Example 1: When the patient (or their family, hereinafter the same) inputs explanatory variables) The patient inputs information on explanatory variables via an application program or website on their patient-side device (e.g., smartphone or PC) (hereinafter collectively referred to as the "patient-side interface"). Based on the input explanatory variables and the palliative care need assessment model MO, the need for palliative care is determined. The resulting assessment (necessity, probability, score, etc.) is displayed on the screen of the patient-side device. The patient reviews the displayed assessment results and, if necessary, sends their preference for palliative care provision and care to the secretariat via the patient-side interface. If the patient prefers in-person consultations, the secretariat matches them with a local cancer center hospital and arranges an appointment. If the patient prefers online consultations, the secretariat matches them with an online palliative care team and arranges an online consultation date.

[0057] (Application Example 2: When the attending physician inputs explanatory variables) The attending physician selects and inputs explanatory variable information from the information in the electronic medical record on the attending physician's device (e.g., smartphone or PC). Alternatively, the device automatically extracts explanatory variable information from the information in the electronic medical record. Based on the input explanatory variables and the palliative care need assessment model MO, the need for palliative care is determined. The obtained assessment result is displayed on the screen of the attending physician's device. The attending physician reviews the displayed assessment result and makes a final decision on whether or not an explanation of palliative care is appropriate. If an explanation of palliative care is appropriate, the attending physician, either personally or by requesting a nurse, provides an explanation of palliative care to the patient according to a dedicated pamphlet, etc. If, as a result of the explanation, the patient wishes to receive palliative care, the attending physician consults with, for example, the hospital's palliative care team. During the consultation, the above assessment result and the patient's wishes are conveyed. For example, if there is no palliative care team in the hospital and permission for online consultation is obtained from the patient, or if there is a palliative care team in the hospital but the patient or family wishes, the request for online palliative care consultation is sent via an application program or website on the attending physician's device.

[0058] (Example 3: When palliative care is provided) If provided by an in-hospital palliative care team (in person), the recommendation is recorded using a template in the electronic medical record. If provided by an external online palliative care team, a report is created through a dedicated system. The report includes, for example, the following: - Physical / medication issues (methods to alleviate physical symptoms, suggestions for medication adjustments) - Mental / spiritual issues (methods to cope with distress, suggestions for medication adjustments) - Cancer diagnosis / treatment issues (sharing information on understanding and choosing diagnosis and treatment, anxieties and worries about side effects of anti-cancer treatment, difficulties in communication with medical professionals) - Social issues (sharing information on economic and social reintegration issues, absence of caregivers, and suggestions for intervention by in-hospital social workers) - Family issues (understanding diagnosis / condition / treatment, suggestions for education on care methods) - Issues regarding place of care (sharing information on the patient's wishes and the family's wishes) - Ethical issues (sharing information on the desire for aggressive treatment and differences in the wishes of the patient and family) - Bereavement issues (sharing information on how bereavement families should respond after the death of a patient and suggestions for how to respond) - Requests for future preparation (providing information on symptoms that are not currently present but are expected to appear)

[0059] (Other) In the above application examples, the primary source of input information (explanatory variables) may be a physician, patient, patient's family, or healthcare professional. There may also be an input screen for necessary items. Furthermore, the output target for the following examples may be a physician, patient, patient's family, or healthcare professional. ・Recommendation to see a doctor: If the score indicating the need for palliative care is high, recommend a visit to a palliative care specialist. ・Increase or decrease in intervention frequency: Suggest increasing or decreasing the intervention frequency according to the above score. For example, if the above score exceeds 50, recommend weekly visits. ・Changes to intervention content: Recommend changes to the intervention content, such as strengthening symptom management or adding psychological support. Interventions include drug therapy for pain management and drug therapy for mental health management. This also includes adding psychological support sessions and recommending rehabilitation to improve quality of life. ・Recommendation of intervention timing: Recommend early introduction of palliative care, or weekly telephone follow-up. ・Sharing of assessment results: Share the assessment results with the patient, physician, healthcare professional, and patient's family and work together to take countermeasures.

[0060] (Modifications) The technologies disclosed herein are not limited to the embodiments described above and can be modified in various forms without departing from the spirit thereof, for example, the following modifications are possible.

[0061] The configuration of the information processing device 100 in the above embodiment is merely an example and can be modified in various ways. Furthermore, the content of the palliative care need determination model acquisition process and the determination process in the above embodiment is merely an example and can be modified in various ways. For example, in the above embodiment, the information processing device 100 acquires the palliative care need determination model MO by generating the palliative care need determination model MO itself, but the information processing device 100 may acquire the palliative care need determination model MO generated by another device.

[0062] In the above embodiment, the explanatory variables used as input to the palliative care need assessment model MO are merely examples and can be modified in various ways. Furthermore, the form of each explanatory variable (continuous, categorical, binary, etc.) can be arbitrarily changed.

[0063] In the above embodiment, at least one of the functional units included in the control unit 110 of the information processing device 100 may be included in another device instead of the control unit 110 of the information processing device 100. The palliative care need determination model acquisition process and the determination process in the above embodiment do not necessarily have to be performed by a single device, but may be performed by separate devices. Each step of the palliative care need determination model acquisition process in the above embodiment does not necessarily have to be performed by a single device, but may be performed by different devices. Similarly, each step of the determination process in the above embodiment does not necessarily have to be performed by a single device, but may be performed by different devices. In the above embodiment, some of the configurations implemented by hardware may be replaced with software, and conversely, some of the configurations implemented by software may be replaced with hardware.

[0064] 100: Information processing unit 110: Control unit 111: Judgment processing unit 112: Original data acquisition unit 113: Training data acquisition unit 114: Model acquisition unit 115: Target data acquisition unit 116: Judgment execution unit 120: Storage unit 130: Display unit 140: Operation input unit 150: Interface unit 190: Bus CP: Judgment processing program LD: Training data MD: Medical record data MO: Palliative care need judgment model RD: Judgment result data

Claims

1. An information processing device comprising: a model acquisition unit that acquires a trained model in which medical record data of a patient with a specific disease before or during a predetermined treatment intervention is used as an explanatory variable and the need for palliative care for the patient is used as the objective variable; a target data acquisition unit that acquires the medical record data of the target patient; and a determination execution unit that inputs the medical record data of the target patient into the trained model to determine the need for palliative care for the target patient and outputs the determination result to an output device.

2. An information processing device according to claim 1, wherein the medical record data includes an index value representing the degree of emotional distress.

3. An information processing device according to claim 1 or claim 2, wherein the medical record data includes an index value representing the degree of pain.

4. An information processing device according to claim 1 or claim 2, wherein the specific disease is cancer.

5. An information processing device according to claim 4, wherein the predetermined treatment is chemotherapy.

6. An information processing device according to claim 1 or claim 2, wherein the determination execution unit, when it is determined in the determination result that there is a high need for palliative care for the target patient, causes a screen for inputting whether or not the provision of palliative care is desired to be displayed on the display device as an output device.

7. An information processing device according to claim 1 or claim 2, wherein the determination execution unit determines the content of recommended palliative care to be provided to the target patient based on the degree of need for palliative care for the target patient in the determination result, and displays the content of the recommended care on a display device as an output device.

8. An information processing device comprising a model acquisition unit that generates a trained model by performing training, using medical record data from before or during a predetermined treatment intervention for a patient with a specific disease as explanatory variables and the need for palliative care for the patient as the dependent variable.

9. An information processing method comprising: a step of obtaining a trained model in which medical record data of a patient with a specific disease before or during a predetermined treatment intervention is used as an explanatory variable and the need for palliative care for the patient is used as the dependent variable; a step of obtaining the medical record data for the target patient; and a step of inputting the medical record data for the target patient into the trained model to determine the need for palliative care for the target patient and outputting the determination result to an output device.

10. An information processing method comprising the step of generating a trained model by performing training, in which medical record data from before or during a predetermined treatment intervention for a patient with a specific disease is used as an explanatory variable, and the need for palliative care for the said patient is used as the dependent variable.

11. A computer program that causes a computer to perform the following steps: acquire a trained model in which medical record data of a patient with a specific disease before or during a predetermined treatment intervention is used as an explanatory variable and the need for palliative care for the patient is used as the dependent variable; acquire the medical record data for the target patient; and input the medical record data for the target patient into the trained model to determine the need for palliative care for the target patient and output the determination result to an output device.

12. A computer program that causes a computer to perform a process of generating a trained model by performing training, with medical record data from before or during a prescribed treatment intervention for a patient with a specific disease as the explanatory variable, and the need for palliative care for the said patient as the dependent variable.

Images