Developing prompts from large language model for extracting clinical information from pathology and ultrasound reports in breast cancer

Purpose@#We aimed to evaluate the time and cost of developing prompts using large language model (LLM), tailored to extract clinical factors in breast cancer patients and their accuracy. @*Materials and Methods@#We collected data from reports of surgical pathology and ultrasound from breast cancer patients who underwent radiotherapy from 2020 to 2022. We extracted the information using the Generative Pre-trained Transformer (GPT) for Sheets and Docs extension plugin and termed this the “LLM” method. The time and cost of developing the prompts with LLM methods were assessed and compared with those spent on collecting information with “full manual” and “LLM-assisted manual” methods. To assess accuracy, 340 patients were randomly selected, and the extracted information by LLM method were compared with those collected by “full manual” method. @*Results@#Data from 2,931 patients were collected. We developed 12 prompts for Extract function and 12 for Format function to extract and standardize the information. The overall accuracy was 87.7%. For lymphovascular invasion, it was 98.2%. Developing and processing the prompts took 3.5 hours and 15 minutes, respectively. Utilizing the ChatGPT application programming interface cost US $65.8 and when factoring in the estimated wage, the total cost was US $95.4. In an estimated comparison, “LLM-assisted manual” and “LLM” methods were time- and cost-efficient compared to the “full manual” method. @*Conclusion@#Developing and facilitating prompts for LLM to derive clinical factors was efficient to extract crucial information from huge medical records. This study demonstrated the potential of the application of natural language processing using LLM model in breast cancer patients. Prompts from the current study can be re-used for other research to collect clinical information.

Quality of Life and Characteristics of Depression with Subjective Cognitive Decline in Korean Adults :Data from the Seventh Korea National Health and Nutrition Examination Survey / 정신신체의학

Objectives@#：This study aimed to investigate quality of life, severity of depression, suicidality, subjective health and subjective stress of depression with subjective cognitive decline in Korean adults. @*Methods@#：We used the 7th KNHANES data to enroll 415 participants with a score of 10 or higher on Patient Health Questionnaire-9 (PHQ-9), aged 20-64. Depression was divided into two groups based on the presence/absence of subjective cognitive decline. Demographic and psychological characteristics were compared between two groups. Correlation analysis of subjective cognitive decline, quality of life, depression, suicidal idea was car-ried out. To detect which variables influenced quality of life, a multiple regression analysis was carried out. @*Results@#：Among the 415 participants, 98 had depression with subjective cognitive decline. We identified sig-nificant differences in age, marital status, education, employment between the two groups. After adjusting for these variables, depression with subjective cognitive decline had lower EuroQol-5D index scores, more severe depressive symptoms without cognition and worse subjective health than depression without cognitive decline. There was a significant correlation between subjective cognitive decline and quality of life (r=-0.236, p<0.001), suicidal idea (r=0.182, p<0.001), depression score without cognition (r=0.108, p=0.028). Through multiple regression analysis, subjective cognitive decline was predictor of reduced quality of life (β=-0.178, p<0.001). @*Conclusions@#：Depression with subjective cognitive decline has poor quality of life and severe depression. Cognitive decline should be considered to improve treatment result in depression.

Comparison of the Suicide Attempt Characteristics Associated with Mixed and Non-Mixed Depression in Koreans / 정신신체의학

Objectives@#：This study aimed to compare the characteristics of suicide attempts among Korean patients with mixed and non-mixed depression. @*Methods@#：Patients who visited the emergency room due to a suicide attempt and participated in the Korean Cohort for the Model Predicting a Suicide and Suicide-related Behavior study were included. Using the Montgomery-Asberg Depression Rating Scale (MADRS) and Young Mania Rating Scale (YMRS), 111 patients were classified into the mixed depression (n=46) and non-mixed depression groups (n=65). The Koukopoulos Mixed Depression Rating Scale (KMDRS) score was calculated using the MADRS and YMRS scores. Suicide attempt characteristics were evaluated using the Columbia Suicidal Severity Rating Scale (C-SSRS) and Suicide Intent Scale (SIS). @*Results@#：In the mixed depression group, the reason item among the ideation intensity score of the C-SSRS was higher, and the deterrent item score was lower. Scores on the timing and suicide note items of the SIS were higher, and scores for overt communication items were lower in the mixed depression group. The KMDRS score was positively correlated with the C-SSRS ideation intensity and total SIS score. After adjusting for additional variables,the KMDRS scores had a significant effect on the C-SSRS ideation intensity and total SIS scores. @*Conclusions@#：The mixed depression group showed a difference in the intensity of suicidal ideation and suicidal intention compared to those in the non-mixed depression group. The overall suicidal ideation intensity and suicidal intention increased according to the degree of mixed depression.

Relationship between Disability Self-Awareness and Insight in Patients with Schizophrenia / 생물치료정신의학

Objectives@#：We aimed to evaluate the relationship between disability self-awareness and insight in patients with schizophrenia. @*Methods@#：We enrolled 58 clinically stable patients with schizophrenia. The World Health Organization Disability Assessment Schedule (WHODAS) self-report was used to identify self-awareness of functional ability. The Positive and Negative Syndrome Scale (PANSS), Clinical Global Impression-Schizophrenia Scale (CGI-SCH), and Personal and Social Performance Scale (PSP) were utilized to evaluate clinical symptoms and global function. Whereas Insight Scale for Psychosis (ISP) was applied for insight. Statistical analyses were performed using correlation and linear regression. @*Results@#：The WHODAS had a significant correlation with the general psychopathology subscale and total score of PANSS, CGI-SCH, and PSP. Moreover, ISP had a strong correlation with the overall WHODAS score and all domains. In the linear regression analysis, ISP had significant effects on the overall WHODAS score after adjusting for additional variables. @*Conclusion@#：Disability self-awareness was significantly correlated with the level of insight in patients with schizophrenia. Clinicians need to consider patients’ insight during the discussion of functional ability.