PURPOSE: We aimed to make a comparison of Boston Questionnaire scores with provocation tests and findings of electrophysiological studies in patients with a carpal tunnel syndrome. MATERIALS AND METHODS: The Boston Questionnaire was applied preoperatively for 248 hands in 142 patients with an idiopathic carpal tunnel syndrome. Boston Questionnaire scores were compared with provocation tests (Phalen's test, Tinel's sign, compression test and hand elevation test) and electrophysiological findings. RESULTS: Correlation study between the Boston Questionnaire scores and electrophysiological findings showed that the electrophysiological findings correlated more with symptom severity score (Spearman coefficient, 0.545; p<0.01) than with functional status score (Spearman coefficient, 0.307; p<0.01). Symptom severity score and functional status scores of the Boston Questionnaire correlated more linearly with the hand elevation test than with other provocation tests. CONCLUSION: This study demonstrates a high correlation of Boston Questionnaire scores with the electrophysiological findings and the hand elevation test.
Boston ; Carpal Tunnel Syndrome ; Hand ; Humans ; Surveys and Questionnaires ; Statistics as Topic

PURPOSE: In IgA nephropathy (IgAN), crescent formation appears to represent a nonspecific response to severe injury to the glomerular capillary wall. This study was performed to evaluate the clinicopathological manifestations of the crescents and their effects on the clinical courses of IgAN. METHODS: The patients diagnosed IgAN were included and the information about their renal biopsies, chemistries and immunohistochemistry findings were collected retrospectively. Some patients that have similar renal function and protenuria were followed up for 12 months to examine the effects of crescents on the renal prognosis. RESULTS: 38 patients with crescents and 177 patients without crescents were enrolled. The patients with IgAN with crescents showed significantly lower renal function (MDRD eGFR 58.5 vs 88.4 ml/min/1.73m2), higher blood pressure, larger amount of proteinuria and more severe hematuria than those patients without crescents. In pathologic findings, HS Lee grades were higher (2.9 vs 1.9). When we selected patients with mildly decreased renal function (serum creatinine <2.5 mg/dL, PCR 0.5-8 g/gCr), the patients with crescents presented lower renal function and higher proteinuria but no statistical significance. After 12 months of treatment, the patients with crescents showed significantly lower renal function (MDRD eGFR 78.6 vs 96.5 ml/min/1.73m2) and higher proteinuria (0.9 vs 0.6 g/gCr). CONCLUSION: The patients with IgAN with crescents showed more deteriorated clinicopathological findings than those without crescents. Despite aggressive treatments, they presented a significantly decreased renal function and larger amount of proteinuria after 1 year. So crescents are supposed to have poor effects on the clinical course.
Biopsy ; Blood Pressure ; Capillaries ; Creatinine ; Glomerulonephritis, IGA ; Hematuria ; Humans ; Immunoglobulin A ; Immunohistochemistry ; Polymerase Chain Reaction ; Proteinuria ; Retrospective Studies

PURPOSE: Adipose-derived stromal cells (ASCs) are readily harvested from lipoaspirated tissue or subcutaneous adipose tissue fragments. The stromal vascular fraction (SVF) is a heterogeneous set of cell populations that surround and support adipose tissue, which includes the stromal cells, ASCs, that have the ability to differentiate into cells of several lineages and contains cells from the microvasculature. The mechanisms that drive the ASCs into the osteoblast lineage are still not clear, but the process has been more extensively studied in bone marrow stromal cells. The purpose of this study was to investigate the osteogenic capacity of adipose derived SVF cells and evaluate bone formation following implantation of SVF cells into the bone defect of human phalanx. METHODS: Case 1 a 43-year-old male was wounded while using a press machine. After first operation, segmental bone defects of the left 3rd and 4th middle phalanx occurred. At first we injected the SVF cells combined with demineralized bone matrix (DBM) to defected 4th middle phalangeal bone lesion. We used P (L/DL)LA [Poly (70L-lactide-co-30DL-lactide) Co Polymer P (L/DL)LA] as a scaffold. Next, we implanted the SVF cells combined with DBM to repair left 3rd middle phalangeal bone defect in sequence. Case 2 was a 25-year-old man with crushing hand injury. Three months after the previous surgery, we implanted the SVF cells combined with DBM to restore right 3rd middle phalangeal bone defect by syringe injection. Radiographic images were taken at follow-up hospital visits and evaluated radiographically by means of computerized analysis of digital images. RESULTS: The phalangeal bone defect was treated with autologous SVF cells isolated and applied in a single operative procedure in combination with DBM. The SVF cells were supported in place with mechanical fixation with a resorbable macroporous sheets acting as a soft tissue barrier. The radiographic appearance of the defect revealed a restoration to average bone density and stable position of pharyngeal bone. Densitometric evaluations for digital X-ray revealed improved bone densities in two cases with pharyngeal bone defects, that is, 65.2% for 4th finger of the case 1, 60.5% for 3rd finger of the case 1 and 60.1% for the case 2. CONCLUSION: This study demonstrated that adipose derived stromal vascular fraction cells have osteogenic potential in two clinical case studies. Thus, these reports show that cells from the SVF cells have potential in many areas of clinical cell therapy and regenerative medicine, albeit a lot of work is yet to be done.
Adipose Tissue ; Adult ; Bone Density ; Bone Matrix ; Durapatite ; Fingers ; Follow-Up Studies ; Hand Injuries ; Humans ; Hypogonadism ; Male ; Mesenchymal Stromal Cells ; Microvessels ; Mitochondrial Diseases ; Ophthalmoplegia ; Osteoblasts ; Osteogenesis ; Polymers ; Regenerative Medicine ; Stromal Cells ; Subcutaneous Fat ; Surgical Procedures, Operative ; Syringes ; Tissue Therapy

A 74-year-old postmenopausal woman visited our gynecology clinic complaining of a palpable abdominal mass. Physical and radiological evaluation indicated that the mass exhibited features of a left ovarian neoplasm showing heterogeneous enhancement. Surgical resection was performed to confirm this suspicion. During surgery, a mass was observed only in the left ovary with no invasive growth, but adhesions to the surrounding peritoneum were seen. Given the patient's age, large mass size, and accompanying uterine myoma and right ovarian cyst, total abdominal hysterectomy with bilateral salpingo-oophorectomy was performed. The final pathologic diagnosis was dedifferentiated liposarcoma. The liposarcoma was suspected to originate from retroperitoneal adipose tissue rather than the ovary. Radiotherapy was planned if a gross lesion indicating recurrence followed 6 months later. This case required a considerable multi-disciplinary approach for diagnosis and treatment because of its ambiguous clinical and radiological findings.
Aged ; Diagnosis ; Female ; Gynecology ; Humans ; Hysterectomy ; Intra-Abdominal Fat ; Leiomyoma ; Liposarcoma* ; Ovarian Cysts ; Ovarian Neoplasms ; Ovary ; Peritoneum ; Radiotherapy ; Recurrence ; Retroperitoneal Neoplasms

Purpose: This study aimed to evaluate the diagnostic accuracy of three multimodal large language models (LLMs) in radiological image interpretation and to assess the impact of prompt engineering strategies and input conditions. Methods: This study analyzed 67 radiological quiz cases from the Korean Society of Ultrasound in Medicine. Three multimodal LLMs (Claude 3.5 Sonnet, GPT-4o, and Gemini-1.5-Pro-002) were evaluated using six types of prompts (basic [without system prompt], original [specific instructions], chain-of-thought, reflection, multiagent, and artificial intelligence [AI]–generated). Performance was assessed across various factors, including tumor versus non-tumor status, case rarity, difficulty, and knowledge cutoff dates. A subgroup analysis compared diagnostic accuracy between imaging-only inputs and combined imaging-descriptive text inputs. Results: With imaging-only inputs, Claude 3.5 Sonnet achieved the highest overall accuracy (46.3%, 186/402), followed by GPT-4o (43.5%, 175/402) and Gemini-1.5-Pro-002 (39.8%, 160/402). AI-generated prompts yielded superior combined accuracy across all three models, with significant improvements over the basic (7.96%, P=0.009), chain-of-thought (6.47%, P=0.029), and multiagent prompts (5.97%, P=0.043). The integration of descriptive text significantly enhanced diagnostic accuracy for Claude 3.5 Sonnet (46.3% to 66.2%, P<0.001), GPT-4o (43.5% to 57.5%, P<0.001), and Gemini-1.5-Pro-002 (39.8% to 60.4%, P<0.001). Model performance was significantly influenced by case rarity for GPT-4o (rare: 6.7% vs. nonrare: 53.9%, P=0.001) and by knowledge cutoff dates for Claude 3.5 Sonnet (post-cutoff: 23.5% vs. pre-cutoff: 64.0%, P=0.005). Conclusion Claude 3.5 Sonnet achieved the highest diagnostic accuracy in radiological quiz cases, followed by GPT-4o and Gemini-1.5-Pro-002. The use of AI-generated prompts and the integration of descriptive text inputs enhanced model performance.

Purpose: This study aimed to evaluate the diagnostic accuracy of three multimodal large language models (LLMs) in radiological image interpretation and to assess the impact of prompt engineering strategies and input conditions. Methods: This study analyzed 67 radiological quiz cases from the Korean Society of Ultrasound in Medicine. Three multimodal LLMs (Claude 3.5 Sonnet, GPT-4o, and Gemini-1.5-Pro-002) were evaluated using six types of prompts (basic [without system prompt], original [specific instructions], chain-of-thought, reflection, multiagent, and artificial intelligence [AI]–generated). Performance was assessed across various factors, including tumor versus non-tumor status, case rarity, difficulty, and knowledge cutoff dates. A subgroup analysis compared diagnostic accuracy between imaging-only inputs and combined imaging-descriptive text inputs. Results: With imaging-only inputs, Claude 3.5 Sonnet achieved the highest overall accuracy (46.3%, 186/402), followed by GPT-4o (43.5%, 175/402) and Gemini-1.5-Pro-002 (39.8%, 160/402). AI-generated prompts yielded superior combined accuracy across all three models, with significant improvements over the basic (7.96%, P=0.009), chain-of-thought (6.47%, P=0.029), and multiagent prompts (5.97%, P=0.043). The integration of descriptive text significantly enhanced diagnostic accuracy for Claude 3.5 Sonnet (46.3% to 66.2%, P<0.001), GPT-4o (43.5% to 57.5%, P<0.001), and Gemini-1.5-Pro-002 (39.8% to 60.4%, P<0.001). Model performance was significantly influenced by case rarity for GPT-4o (rare: 6.7% vs. nonrare: 53.9%, P=0.001) and by knowledge cutoff dates for Claude 3.5 Sonnet (post-cutoff: 23.5% vs. pre-cutoff: 64.0%, P=0.005). Conclusion Claude 3.5 Sonnet achieved the highest diagnostic accuracy in radiological quiz cases, followed by GPT-4o and Gemini-1.5-Pro-002. The use of AI-generated prompts and the integration of descriptive text inputs enhanced model performance.

Purpose: This study aimed to evaluate the diagnostic accuracy of three multimodal large language models (LLMs) in radiological image interpretation and to assess the impact of prompt engineering strategies and input conditions. Methods: This study analyzed 67 radiological quiz cases from the Korean Society of Ultrasound in Medicine. Three multimodal LLMs (Claude 3.5 Sonnet, GPT-4o, and Gemini-1.5-Pro-002) were evaluated using six types of prompts (basic [without system prompt], original [specific instructions], chain-of-thought, reflection, multiagent, and artificial intelligence [AI]–generated). Performance was assessed across various factors, including tumor versus non-tumor status, case rarity, difficulty, and knowledge cutoff dates. A subgroup analysis compared diagnostic accuracy between imaging-only inputs and combined imaging-descriptive text inputs. Results: With imaging-only inputs, Claude 3.5 Sonnet achieved the highest overall accuracy (46.3%, 186/402), followed by GPT-4o (43.5%, 175/402) and Gemini-1.5-Pro-002 (39.8%, 160/402). AI-generated prompts yielded superior combined accuracy across all three models, with significant improvements over the basic (7.96%, P=0.009), chain-of-thought (6.47%, P=0.029), and multiagent prompts (5.97%, P=0.043). The integration of descriptive text significantly enhanced diagnostic accuracy for Claude 3.5 Sonnet (46.3% to 66.2%, P<0.001), GPT-4o (43.5% to 57.5%, P<0.001), and Gemini-1.5-Pro-002 (39.8% to 60.4%, P<0.001). Model performance was significantly influenced by case rarity for GPT-4o (rare: 6.7% vs. nonrare: 53.9%, P=0.001) and by knowledge cutoff dates for Claude 3.5 Sonnet (post-cutoff: 23.5% vs. pre-cutoff: 64.0%, P=0.005). Conclusion Claude 3.5 Sonnet achieved the highest diagnostic accuracy in radiological quiz cases, followed by GPT-4o and Gemini-1.5-Pro-002. The use of AI-generated prompts and the integration of descriptive text inputs enhanced model performance.

Purpose: This study aimed to evaluate the diagnostic accuracy of three multimodal large language models (LLMs) in radiological image interpretation and to assess the impact of prompt engineering strategies and input conditions. Methods: This study analyzed 67 radiological quiz cases from the Korean Society of Ultrasound in Medicine. Three multimodal LLMs (Claude 3.5 Sonnet, GPT-4o, and Gemini-1.5-Pro-002) were evaluated using six types of prompts (basic [without system prompt], original [specific instructions], chain-of-thought, reflection, multiagent, and artificial intelligence [AI]–generated). Performance was assessed across various factors, including tumor versus non-tumor status, case rarity, difficulty, and knowledge cutoff dates. A subgroup analysis compared diagnostic accuracy between imaging-only inputs and combined imaging-descriptive text inputs. Results: With imaging-only inputs, Claude 3.5 Sonnet achieved the highest overall accuracy (46.3%, 186/402), followed by GPT-4o (43.5%, 175/402) and Gemini-1.5-Pro-002 (39.8%, 160/402). AI-generated prompts yielded superior combined accuracy across all three models, with significant improvements over the basic (7.96%, P=0.009), chain-of-thought (6.47%, P=0.029), and multiagent prompts (5.97%, P=0.043). The integration of descriptive text significantly enhanced diagnostic accuracy for Claude 3.5 Sonnet (46.3% to 66.2%, P<0.001), GPT-4o (43.5% to 57.5%, P<0.001), and Gemini-1.5-Pro-002 (39.8% to 60.4%, P<0.001). Model performance was significantly influenced by case rarity for GPT-4o (rare: 6.7% vs. nonrare: 53.9%, P=0.001) and by knowledge cutoff dates for Claude 3.5 Sonnet (post-cutoff: 23.5% vs. pre-cutoff: 64.0%, P=0.005). Conclusion Claude 3.5 Sonnet achieved the highest diagnostic accuracy in radiological quiz cases, followed by GPT-4o and Gemini-1.5-Pro-002. The use of AI-generated prompts and the integration of descriptive text inputs enhanced model performance.

Purpose: This study aimed to evaluate the diagnostic accuracy of three multimodal large language models (LLMs) in radiological image interpretation and to assess the impact of prompt engineering strategies and input conditions. Methods: This study analyzed 67 radiological quiz cases from the Korean Society of Ultrasound in Medicine. Three multimodal LLMs (Claude 3.5 Sonnet, GPT-4o, and Gemini-1.5-Pro-002) were evaluated using six types of prompts (basic [without system prompt], original [specific instructions], chain-of-thought, reflection, multiagent, and artificial intelligence [AI]–generated). Performance was assessed across various factors, including tumor versus non-tumor status, case rarity, difficulty, and knowledge cutoff dates. A subgroup analysis compared diagnostic accuracy between imaging-only inputs and combined imaging-descriptive text inputs. Results: With imaging-only inputs, Claude 3.5 Sonnet achieved the highest overall accuracy (46.3%, 186/402), followed by GPT-4o (43.5%, 175/402) and Gemini-1.5-Pro-002 (39.8%, 160/402). AI-generated prompts yielded superior combined accuracy across all three models, with significant improvements over the basic (7.96%, P=0.009), chain-of-thought (6.47%, P=0.029), and multiagent prompts (5.97%, P=0.043). The integration of descriptive text significantly enhanced diagnostic accuracy for Claude 3.5 Sonnet (46.3% to 66.2%, P<0.001), GPT-4o (43.5% to 57.5%, P<0.001), and Gemini-1.5-Pro-002 (39.8% to 60.4%, P<0.001). Model performance was significantly influenced by case rarity for GPT-4o (rare: 6.7% vs. nonrare: 53.9%, P=0.001) and by knowledge cutoff dates for Claude 3.5 Sonnet (post-cutoff: 23.5% vs. pre-cutoff: 64.0%, P=0.005). Conclusion Claude 3.5 Sonnet achieved the highest diagnostic accuracy in radiological quiz cases, followed by GPT-4o and Gemini-1.5-Pro-002. The use of AI-generated prompts and the integration of descriptive text inputs enhanced model performance.

Monocyte chemoattractant protein-1 (MCP-1) is suggested to be involved in the progression of diabetic nephropathy. We investigated the association of the -2518 A/G polymorphism in the MCP-1 gene with progressive kidney failure in Korean patients with type 2 diabetes mellitus (DM). We investigated -2518 A/G polymorphism of the MCP-1 gene in type 2 DM patients with progressive kidney failure (n=112) compared with matched type 2 DM patients without nephropathy (diabetic control, n=112) and healthy controls (n=230). The overall genotypic distribution of -2518 A/G in the MCP-1 gene was not different in patients with type 2 DM compared to healthy controls. Although the genotype was not significantly different between the patients with kidney failure and the diabetic control (p=0.07), the A allele was more frequent in patients with kidney failure than in DM controls (42.0 vs. 32.1%, p=0.03). The carriage of A allele was significantly associated with kidney failure (68.8 vs. 54.5%, OR 1.84, 95% CI 1.07-3.18). In logistic regression analysis, carriage of A allele retained a significant association with diabetic kidney failure. Our result shows that the -2518 A allele of the MCP-1 gene is associated with kidney failure in Korean patients with type 2 DM.
Adult ; Aged ; Alleles ; Chemokine CCL2/*metabolism ; Diabetes Mellitus, Type 2/ethnology/*genetics/*metabolism ; Diabetic Nephropathies/ethnology/*genetics/*metabolism ; Female ; Genotype ; Humans ; Kidney Failure ; Korea ; Male ; Middle Aged ; *Polymorphism, Genetic ; *Promoter Regions, Genetic ; Risk Factors