Emergency Patient Triage Improvement through a Retrieval-Augmented Generation Enhanced Large-Scale Language Model.

OBJECTIVES: Emergency medical triage is crucial for prioritizing patient care in emergency situations, yet its effectiveness can vary significantly based on the experience and training of the personnel involved. This study aims to evaluate the efficacy of integrating Retrieval Augmented Generation (RAG) with Large Language Models (LLMs), specifically OpenAI's GPT models, to standardize triage procedures and reduce variability in emergency care. METHODS: We created 100 simulated triage scenarios based on modified cases from the Japanese National Examination for Emergency Medical Technicians. These scenarios were processed by the RAG-enhanced LLMs, and the models were given patient vital signs, symptoms, and observations from emergency medical services (EMS) teams as inputs. The primary outcome was the accuracy of triage classifications, which was used to compare the performance of the RAG-enhanced LLMs with that of emergency medical technicians and emergency physicians. Secondary outcomes included the rates of under-triage and over-triage. RESULTS: The Generative Pre-trained Transformer 3.5 (GPT-3.5) with RAG model achieved a correct triage rate of 70%, significantly outperforming Emergency Medical Technicians (EMTs) with 35% and 38% correct rates, and emergency physicians with 50% and 47% correct rates (p < 0.05). Additionally, this model demonstrated a substantial reduction in under-triage rates to 8%, compared with 33% for GPT-3.5 without RAG, and 39% for GPT-4 without RAG. CONCLUSIONS: The integration of RAG with LLMs shows promise in improving the accuracy and consistency of medical assessments in emergency settings. Further validation in diverse medical settings with broader datasets is necessary to confirm the effectiveness and adaptability of these technologies in live environments.

Invasive infection due to Streptococcus dysgalactiae subspecies equisimilis causing endocarditis and ventriculitis: A case report.

We aim to raise awareness of the role of Streptococcus dysgalactiae subsp. equisimilis (SDSE) in causing endovascular and central nervous system infections, and to promote recognition of SDSE as a pathogen that may cause serious invasive infections.

Urosepsis secondary to ureterosciatic hernia corrected with ureteral stent placement: a case report and literature review.

BACKGROUND: Ureterosciatic hernia is a rare type of pelvic floor herniation that occurs through the sciatic foramen. The resulting ureteral obstruction may lead to hydronephrosis and to further complications including urinary tract infection and urosepsis. There have been 30 reported cases of ureterosciatic hernia. Ureteral stenting and surgical repair have been used as treatment options. CASE PRESENTATION: We report the case of an 86-year-old woman who was transferred to Tokyo Metropolitan Bokutoh Hospital with symptoms of fever and septic shock. Her computed tomography scan revealed left hydronephrosis and deviation of the left ureter into the sciatic foramen; she was therefore diagnosed with a left ureteral sciatic hernia and admitted in our intensive care unit for further treatment with resuscitative fluids, vasopressors, and antibiotics. Following a retrograde insertion ureteral catheter insertion, ureteral incarceration was relieved, and a double-J ureteral stent was placed in situ. Antibiotic treatment was initiated, and the patient's hemodynamic status gradually improved. CONCLUSIONS: Although ureterosciatic hernia is a rare disorder, it is associated with serious complications including urinary tract infection with sepsis, which may warrant urgent corrective procedure to relieve the structural obstruction. Treatment may be conservative or surgical, though treatment with ureteral stent placement may be a favorable approach in elderly patients with multiple comorbidities presenting with urosepsis.

Melanoma antigen A12 regulates cell cycle via tumor suppressor p21 expression.

Melanoma-associated antigen family A (MAGE-A) is a family of cancer/testis antigens that are expressed in malignant tumors but not in normal tissues other than the testes. MAGE-A12 is a MAGE-A family gene whose tumorigenic function in cancer cells remains unclear. Searches of the Oncomine and NextBio databases revealed that malignant tumors show up-regulation of MAGE-A12 mRNA relative to corresponding normal tissue. In PPC1 primary prostatic carcinoma cells and in HCT116 colorectal cancer cells (wild type and p53-depleted), MAGE-A12 gene knockdown using siRNA or shRNA diminishes cancer cell proliferation as assessed by cellular ATP levels, cell counting, and clonogenic assays. FACS analyses of annexin V-PI staining and DNA content show that MAGE-A12 knockdown causes G2/M arrest and apoptosis. In tumor xenografts of HCT116 cells, conditional knockdown of MAGE-A12 suppresses tumor growth. The depletion of MAGE-A12 leads to the accumulation of tumor suppressor p21 in PPC1, HCT116, and p53-depleted HCT116 cells. Conversely, CDKN1A knockdown partially rescues the viability of PPC1 cells transfected with siRNA targeting MAGE-A12, while p21 overexpression leads to proliferation arrest in PPC-1 cells. Furthermore, exogenous MAGE-A12 expression promotes the ubiquitination of p21. Our findings reveal that MAGE-A12 plays crucial roles in p21 stability and tumor growth, suggesting that MAGE-A12 could provide a novel target for cancer treatment.

Lipid Nanoparticle-mediated siRNA Transfer Against PCTAIRE1/PCTK1/Cdk16 Inhibits In Vivo Cancer Growth.

PCTAIRE1/CDK16/PCTK1 plays critical roles in cancer cell proliferation and antiapoptosis. To advance our previously published in vitro results with PCTAIRE1 silencing, we examined the in vivo therapeutic potential of this approach by using small interfering RNA (siRNA) encapsulated by lipid nanoparticles. Therapy experiments of PCTAIRE1 siRNA were performed using human HCT116 colorectal cancer cells and human A2058 melanoma cells. A single dose of PCTAIRE1 siRNA-lipid nanoparticles was found to be highly effective in reducing in vivo PCTAIRE1 expression for up to 4 days as assayed by immunoblotting. Therapy experiments were started 4 days after subcutaneous injection of cancer cells. Treatment with PCTAIRE1 siRNA-lipid nanoparticles (0.5 mg/kg RNA, twice a week) reduced tumor volume and weight significantly compared with the scramble-control group. Histopathological analysis (terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling) showed increased apoptosis of tumor cells treated with PCTAIRE1-siRNA. Overall, our results demonstrate that siRNA treatment targeting PCTAIRE1 is effective in vivo, suggesting that PCTAIRE1 siRNA-lipid nanoparticles might be a novel therapeutic approach against cancer cells.