Background: Idiopathic pulmonary fibrosis (IPF) is a chronic, progressive lung disease that culminates in respiratory failure and death due to irreversible scarring of the distal lung. While initially considered a chronic inflammatory disorder, the aberrant function of the alveolar epithelium is now acknowledged as playing a central role in the pathophysiology of IPF. This study aimed to investigate the regenerative capacity of alveolar type 2 (AT2) cells using IPF-derived alveolar organoids and to examine the effects of disease progression on this capacity. Methods: Lung tissues from three pneumothorax patients and six IPF patients (early and advanced stages) were obtained through video-assisted thoracoscopic surgery and lung transplantation. HTII-280+ cells were isolated from CD31-CD45-epithelial cell adhesion molecule (EpCAM)+ cells in the distal lungs of IPF and pneumothorax patients using fluorescence-activated cell sorting (FACS) and resuspended in 48-well plates to establish IPF-derived alveolar organoids. Immunostaining was used to verify the presence of AT2 cells. Results: FACS sorting yielded approximately 1% of AT2 cells in early IPF tissue, and the number decreased as the disease progressed, in contrast to 2.7% in pneumothorax. Additionally, the cultured organoids in the IPF groups were smaller and less numerous compared to those from pneumothorax patients. The colony forming efficiency decreased as the disease advanced. Immunostaining results showed that the IPF organoids expressed less surfactant protein C (SFTPC) compared to the pneumothorax group and contained keratin 5+ (KRT5+) cells. Conclusion This study confirmed that the regenerative capacity of AT2 cells in IPF decreases as the disease progresses, with IPF-derived AT2 cells inherently exhibiting functional abnormalities and altered differentiation plasticity.

Objectives: Developing large language models (LLMs) in biomedicine requires access to high-quality training and alignment tuning datasets. However, publicly available Korean medical preference datasets are scarce, hindering the advancement of Korean medical LLMs. This study constructs and evaluates the efficacy of the Korean Medical Preference Dataset (KoMeP), an alignment tuning dataset constructed with an automated pipeline, minimizing the high costs of human annotation. Methods: KoMeP was generated using the DAHL score, an automated hallucination evaluation metric. Five LLMs (Dolly-v2-3B, MPT-7B, GPT-4o, Qwen-2-7B, Llama-3-8B) produced responses to 8,573 biomedical examination questions, from which 5,551 preference pairs were extracted. Each pair consisted of a “chosen” response and a “rejected” response, as determined by their DAHL scores. The dataset was evaluated when trained through two different alignment tuning methods, direct preference optimization (DPO) and odds ratio preference optimization (ORPO) respectively across five different models. The KorMedMCQA benchmark was employed to assess the effectiveness of alignment tuning. Results: Models trained with DPO consistently improved KorMedMCQA performance; notably, Llama-3.1-8B showed a 43.96% increase. In contrast, ORPO training produced inconsistent results. Additionally, English-to-Korean transfer learning proved effective, particularly for English-centric models like Gemma-2, whereas Korean-to-English transfer learning achieved limited success. Instruction tuning with KoMeP yielded mixed outcomes, which suggests challenges in dataset formatting. Conclusions KoMeP is the first publicly available Korean medical preference dataset and significantly improves alignment tuning performance in LLMs. The DPO method outperforms ORPO in alignment tuning. Future work should focus on expanding KoMeP, developing a Korean-native dataset, and refining alignment tuning methods to produce safer and more reliable Korean medical LLMs.

Background: Idiopathic pulmonary fibrosis (IPF) is a chronic, progressive lung disease that culminates in respiratory failure and death due to irreversible scarring of the distal lung. While initially considered a chronic inflammatory disorder, the aberrant function of the alveolar epithelium is now acknowledged as playing a central role in the pathophysiology of IPF. This study aimed to investigate the regenerative capacity of alveolar type 2 (AT2) cells using IPF-derived alveolar organoids and to examine the effects of disease progression on this capacity. Methods: Lung tissues from three pneumothorax patients and six IPF patients (early and advanced stages) were obtained through video-assisted thoracoscopic surgery and lung transplantation. HTII-280+ cells were isolated from CD31-CD45-epithelial cell adhesion molecule (EpCAM)+ cells in the distal lungs of IPF and pneumothorax patients using fluorescence-activated cell sorting (FACS) and resuspended in 48-well plates to establish IPF-derived alveolar organoids. Immunostaining was used to verify the presence of AT2 cells. Results: FACS sorting yielded approximately 1% of AT2 cells in early IPF tissue, and the number decreased as the disease progressed, in contrast to 2.7% in pneumothorax. Additionally, the cultured organoids in the IPF groups were smaller and less numerous compared to those from pneumothorax patients. The colony forming efficiency decreased as the disease advanced. Immunostaining results showed that the IPF organoids expressed less surfactant protein C (SFTPC) compared to the pneumothorax group and contained keratin 5+ (KRT5+) cells. Conclusion This study confirmed that the regenerative capacity of AT2 cells in IPF decreases as the disease progresses, with IPF-derived AT2 cells inherently exhibiting functional abnormalities and altered differentiation plasticity.

Background: Idiopathic pulmonary fibrosis (IPF) is a chronic, progressive lung disease that culminates in respiratory failure and death due to irreversible scarring of the distal lung. While initially considered a chronic inflammatory disorder, the aberrant function of the alveolar epithelium is now acknowledged as playing a central role in the pathophysiology of IPF. This study aimed to investigate the regenerative capacity of alveolar type 2 (AT2) cells using IPF-derived alveolar organoids and to examine the effects of disease progression on this capacity. Methods: Lung tissues from three pneumothorax patients and six IPF patients (early and advanced stages) were obtained through video-assisted thoracoscopic surgery and lung transplantation. HTII-280+ cells were isolated from CD31-CD45-epithelial cell adhesion molecule (EpCAM)+ cells in the distal lungs of IPF and pneumothorax patients using fluorescence-activated cell sorting (FACS) and resuspended in 48-well plates to establish IPF-derived alveolar organoids. Immunostaining was used to verify the presence of AT2 cells. Results: FACS sorting yielded approximately 1% of AT2 cells in early IPF tissue, and the number decreased as the disease progressed, in contrast to 2.7% in pneumothorax. Additionally, the cultured organoids in the IPF groups were smaller and less numerous compared to those from pneumothorax patients. The colony forming efficiency decreased as the disease advanced. Immunostaining results showed that the IPF organoids expressed less surfactant protein C (SFTPC) compared to the pneumothorax group and contained keratin 5+ (KRT5+) cells. Conclusion This study confirmed that the regenerative capacity of AT2 cells in IPF decreases as the disease progresses, with IPF-derived AT2 cells inherently exhibiting functional abnormalities and altered differentiation plasticity.

Objectives: Developing large language models (LLMs) in biomedicine requires access to high-quality training and alignment tuning datasets. However, publicly available Korean medical preference datasets are scarce, hindering the advancement of Korean medical LLMs. This study constructs and evaluates the efficacy of the Korean Medical Preference Dataset (KoMeP), an alignment tuning dataset constructed with an automated pipeline, minimizing the high costs of human annotation. Methods: KoMeP was generated using the DAHL score, an automated hallucination evaluation metric. Five LLMs (Dolly-v2-3B, MPT-7B, GPT-4o, Qwen-2-7B, Llama-3-8B) produced responses to 8,573 biomedical examination questions, from which 5,551 preference pairs were extracted. Each pair consisted of a “chosen” response and a “rejected” response, as determined by their DAHL scores. The dataset was evaluated when trained through two different alignment tuning methods, direct preference optimization (DPO) and odds ratio preference optimization (ORPO) respectively across five different models. The KorMedMCQA benchmark was employed to assess the effectiveness of alignment tuning. Results: Models trained with DPO consistently improved KorMedMCQA performance; notably, Llama-3.1-8B showed a 43.96% increase. In contrast, ORPO training produced inconsistent results. Additionally, English-to-Korean transfer learning proved effective, particularly for English-centric models like Gemma-2, whereas Korean-to-English transfer learning achieved limited success. Instruction tuning with KoMeP yielded mixed outcomes, which suggests challenges in dataset formatting. Conclusions KoMeP is the first publicly available Korean medical preference dataset and significantly improves alignment tuning performance in LLMs. The DPO method outperforms ORPO in alignment tuning. Future work should focus on expanding KoMeP, developing a Korean-native dataset, and refining alignment tuning methods to produce safer and more reliable Korean medical LLMs.

Background: Idiopathic pulmonary fibrosis (IPF) is a chronic, progressive lung disease that culminates in respiratory failure and death due to irreversible scarring of the distal lung. While initially considered a chronic inflammatory disorder, the aberrant function of the alveolar epithelium is now acknowledged as playing a central role in the pathophysiology of IPF. This study aimed to investigate the regenerative capacity of alveolar type 2 (AT2) cells using IPF-derived alveolar organoids and to examine the effects of disease progression on this capacity. Methods: Lung tissues from three pneumothorax patients and six IPF patients (early and advanced stages) were obtained through video-assisted thoracoscopic surgery and lung transplantation. HTII-280+ cells were isolated from CD31-CD45-epithelial cell adhesion molecule (EpCAM)+ cells in the distal lungs of IPF and pneumothorax patients using fluorescence-activated cell sorting (FACS) and resuspended in 48-well plates to establish IPF-derived alveolar organoids. Immunostaining was used to verify the presence of AT2 cells. Results: FACS sorting yielded approximately 1% of AT2 cells in early IPF tissue, and the number decreased as the disease progressed, in contrast to 2.7% in pneumothorax. Additionally, the cultured organoids in the IPF groups were smaller and less numerous compared to those from pneumothorax patients. The colony forming efficiency decreased as the disease advanced. Immunostaining results showed that the IPF organoids expressed less surfactant protein C (SFTPC) compared to the pneumothorax group and contained keratin 5+ (KRT5+) cells. Conclusion This study confirmed that the regenerative capacity of AT2 cells in IPF decreases as the disease progresses, with IPF-derived AT2 cells inherently exhibiting functional abnormalities and altered differentiation plasticity.

Objectives: Developing large language models (LLMs) in biomedicine requires access to high-quality training and alignment tuning datasets. However, publicly available Korean medical preference datasets are scarce, hindering the advancement of Korean medical LLMs. This study constructs and evaluates the efficacy of the Korean Medical Preference Dataset (KoMeP), an alignment tuning dataset constructed with an automated pipeline, minimizing the high costs of human annotation. Methods: KoMeP was generated using the DAHL score, an automated hallucination evaluation metric. Five LLMs (Dolly-v2-3B, MPT-7B, GPT-4o, Qwen-2-7B, Llama-3-8B) produced responses to 8,573 biomedical examination questions, from which 5,551 preference pairs were extracted. Each pair consisted of a “chosen” response and a “rejected” response, as determined by their DAHL scores. The dataset was evaluated when trained through two different alignment tuning methods, direct preference optimization (DPO) and odds ratio preference optimization (ORPO) respectively across five different models. The KorMedMCQA benchmark was employed to assess the effectiveness of alignment tuning. Results: Models trained with DPO consistently improved KorMedMCQA performance; notably, Llama-3.1-8B showed a 43.96% increase. In contrast, ORPO training produced inconsistent results. Additionally, English-to-Korean transfer learning proved effective, particularly for English-centric models like Gemma-2, whereas Korean-to-English transfer learning achieved limited success. Instruction tuning with KoMeP yielded mixed outcomes, which suggests challenges in dataset formatting. Conclusions KoMeP is the first publicly available Korean medical preference dataset and significantly improves alignment tuning performance in LLMs. The DPO method outperforms ORPO in alignment tuning. Future work should focus on expanding KoMeP, developing a Korean-native dataset, and refining alignment tuning methods to produce safer and more reliable Korean medical LLMs.