OBJECTIVE: Traditional knowledge-based and machine learning diagnostic decision support systems have benefited from integrating the medical domain knowledge encoded in the Unified Medical Language System (UMLS). The emergence of Large Language Models (LLMs) to supplant traditional systems poses questions of the quality and extent of the medical knowledge in the models\' internal knowledge representations and the need for external knowledge sources. The objective of this study is three-fold: to probe the diagnosis-related medical knowledge of popular LLMs, to examine the benefit of providing the UMLS knowledge to LLMs (grounding the diagnosis predictions), and to evaluate the correlations between human judgments and the UMLS-based metrics for generations by LLMs.
METHODS: We evaluated diagnoses generated by LLMs from consumer health questions and daily care notes in the electronic health records using the ConsumerQA and Problem Summarization datasets. Probing LLMs for the UMLS knowledge was performed by prompting the LLM to complete the diagnosis-related UMLS knowledge paths. Grounding the predictions was examined in an approach that integrated the UMLS graph paths and clinical notes in prompting the LLMs. The results were compared to prompting without the UMLS paths. The final experiments examined the alignment of different evaluation metrics, UMLS-based and non-UMLS, with human expert evaluation.
RESULTS: In probing the UMLS knowledge, GPT-3.5 significantly outperformed Llama2 and a simple baseline yielding an F1 score of 10.9% in completing one-hop UMLS paths for a given concept. Grounding diagnosis predictions with the UMLS paths improved the results for both models on both tasks, with the highest improvement (4%) in SapBERT score. There was a weak correlation between the widely used evaluation metrics (ROUGE and SapBERT) and human judgments.
CONCLUSIONS: We found that while popular LLMs contain some medical knowledge in their internal representations, augmentation with the UMLS knowledge provides performance gains around diagnosis generation. The UMLS needs to be tailored for the task to improve the LLMs predictions. Finding evaluation metrics that are aligned with human judgments better than the traditional ROUGE and BERT-based scores remains an open research question.

Diabetic lumbosacral radiculoplexus neuropathy (DLSRPN), also known as diabetic amyotrophy, is a rare disease of exclusion that is difficult to diagnose due to its non-specific clinical presentation of neuropathy, autonomic symptoms, and potential weight loss. Due to this, many differential diagnoses are raised before making a diagnosis of such an uncommon disease. However, once the diagnosis is made, the management of this disease can vary. Here, we would like to discuss the etiology, pathophysiology, diagnosis, and management of this disease, as well as present a rare case of diabetic lumbosacral radiculoplexus neuropathy in a 50-year-old male.

Artificial intelligence (AI) is a suite of technologies that enables computers to learn and interpret information like human cognition. It has found applications across various fields, including healthcare, agriculture, astronomy, navigation, and robotics. Within healthcare, AI has the potential to enhance diagnostic accuracy, facilitate drug research, and automate patient experiences. This comparative study focuses on the proficiency of AI in generating accurate differential diagnoses in the field of pathology. Six medical vignettes were crafted, and each scenario was then input into three different AI platforms. The pathologist reviewed and determined the most accurate AI model.

Neuralgic muscular atrophy is not uncommon in clinical practice. Due to the different branches of brachial plexus involved in the lesion, the clinical symptoms are different, and there is a lack of clear imaging diagnostic criteria, so the diagnosis of this disease brings great challenges to clinicians. We have certain experience in the diagnosis and treatment of this disease, and hereby select a representative case of neuralgic muscular atrophy to share its diagnosis and treatment process, focusing on analyzing the characteristic symptoms of this disease, valuable imaging data and targeted treatment, so as to enable clinicians to better understand this disease.

OBJECTIVE: Neurosurgery emphasizes the criticality of accurate differential diagnoses, with diagnostic delays posing significant health and economic challenges. As large language models (LLMs) emerge as transformative tools in healthcare, this study seeks to elucidate their role in assisting neurosurgeons with the differential diagnosis process, especially during preliminary consultations.
METHODS: This study employed 3 chat-based LLMs, ChatGPT (versions 3.5 and 4.0), Perplexity AI, and Bard AI, to evaluate their diagnostic accuracy. Each LLM was prompted using clinical vignettes, and their responses were recorded to generate differential diagnoses for 20 common and uncommon neurosurgical disorders. Disease-specific prompts were crafted using Dynamed, a clinical reference tool. The accuracy of the LLMs was determined based on their ability to identify the target disease within their top differential diagnoses correctly.
RESULTS: For the initial differential, ChatGPT 3.5 achieved an accuracy of 52.63%, while ChatGPT 4.0 performed slightly better at 53.68%. Perplexity AI and Bard AI demonstrated 40.00% and 29.47% accuracy, respectively. As the number of considered differentials increased from 2 to 5, ChatGPT 3.5 reached its peak accuracy of 77.89% for the top 5 differentials. Bard AI and Perplexity AI had varied performances, with Bard AI improving in the top 5 differentials at 62.11%. On a disease-specific note, the LLMs excelled in diagnosing conditions like epilepsy and cervical spine stenosis but faced challenges with more complex diseases such as Moyamoya disease and amyotrophic lateral sclerosis.
CONCLUSIONS: LLMs showcase the potential to enhance diagnostic accuracy and decrease the incidence of missed diagnoses in neurosurgery.

UNASSIGNED: Autosomal dominant polycystic kidney disease (ADPKD) is the most common hereditary disease causing chronic renal failure, with a high incidence of extra-renal manifestations including pericardial effusion.
UNASSIGNED: We present the case of a 41-year-old female, known for ADPKD, who presented to our emergency department with epigastric pain radiating to the interscapular area. Blood exams showed moderate increase in inflammatory markers. Echocardiography revealed a circumferential pericardial effusion of 10 mm. She was put under treatment with colchicine therapy (1 mg b.i.d.) based on a presumptive diagnosis of acute pericarditis with pericardial effusion. She was hospitalized due to increase in pericardial effusion, underwent pericardial drainage, and started prednisone therapy with rapid recovery. We started a close follow-up on a monthly basis, with progressive decrease in pericardial effusion and progressive amelioration in symptoms, although the patient continued to report mild asthenia.
UNASSIGNED: Pericardial effusion and ADPKD are conditions that both require an interdisciplinary discussion for optimal patient care that avoids neglecting pivotal symptoms and avoidable invasive examinations.

Subacute thyroiditis (SAT) is an inflammatory disease of the thyroid gland. It can present with overlapping features of other etiology of thyroiditis. It can present with thyroid enlargement and systemic symptoms such as fever as well as neck pain and may be confused with infectious thyroiditis. It can be difficult to diagnose and present as fever of unknown origin (FUO). A good history, physical examination, laboratory investigation, as well as imaging may aid in the correct diagnosis and prevent the inappropriate use of antibiotics. Treatment is usually with nonsteroidal anti-inflammatory drugs (NSAIDs) as well as corticosteroids. We herein present a case of SAT presenting as FUO. We highlighted the importance of proper clinical evaluation, the importance of thyroid imaging, and how to differentiate other forms of thyroiditis.

Splenogonadal fusion is a rare, benign congenital malformation characterized by the association of splenic tissue and gonads (typically testicles). It is a condition of male predominance and can be classified into two types: continuous, if the spleen and gonad are united by a splenic cord or fibrous tissue, or discontinuous. Splenogonadal fusion is often associated with other congenital anomalies such as cryptorchidism, limb defects, and micrognathia. Differential diagnosis can be difficult and includes inguinal hernia, spermatic cord cyst, cryptorchidism, or testicular mass. Due to little knowledge of the pathology, unnecessary orchidectomies are often performed. A previously healthy five-year-old boy was sent to a pediatric surgery appointment due to testicular asymmetry. The physical examination showed a painless, nodular mass adhering to the upper pole of the left testicle, without any palpable inguinal masses. Tumor markers were negative, and a testicular ultrasound with Doppler revealed a mass suggestive of an accessory testicle. Left inguinal surgical exploration revealed the presence of a mass joined by fibrous tissue to the upper pole of the testicle, but no connection to the native spleen was found. Total excision was performed with the testicle\'s preservation. The anatomopathological analysis revealed morphological aspects compatible with splenic tissue with normal characteristics. The diagnosis of splenogonadal fusion is rare and complex, requires several differential diagnoses, and is often made intraoperatively.The prognosis is excellent as long as there are no associated malformations. A high level of suspicion for this pathology, with recognition of the anatomical structures, can avoid unnecessary orchidectomy.

BACKGROUND: Migraine, a frequent and debilitating neurological disease, shows gender-specific differences in prevalence and severity. Pregnancy is associated with numerous unique features in terms of migraine course, treatment options and differential diagnoses.
OBJECTIVE: How does pregnancy influence the course of migraine? What are the possible treatment options during pregnancy? Which differential diagnoses should be considered?
METHODS: Narrative review with summary and discussion of relevant studies and guidelines on migraine in pregnancy.
RESULTS: During pregnancy up to three quarters of women experience improvement of their migraine; however, there may be a renewed increase in frequency after childbirth. Choosing an appropriate treatment during pregnancy requires a careful risk-benefit assessment. It is important to consider secondary causes of headache as these can occur more frequently during pregnancy and some can be life-threatening.
CONCLUSIONS: Consideration of specific aspects of migraine in pregnancy is crucial to be able to develop the best possible treatment strategies for affected patients.
UNASSIGNED: HINTERGRUND: Migräne, eine häufige und beeinträchtigende neurologische Erkrankung, zeigt geschlechtsspezifische Unterschiede in Prävalenz und Schweregrad. Die Schwangerschaft geht mit zahlreichen Besonderheiten in Bezug auf Migräneverlauf, Therapieoptionen und Differenzialdiagnosen einher.
UNASSIGNED: Wie beeinflusst eine Schwangerschaft den Migräneverlauf? Was sind mögliche Therapieoptionen während der Schwangerschaft? Welche Differenzialdiagnosen sollten in Betracht gezogen werden?
METHODS: Narratives Review mit Zusammenfassung und Diskussion relevanter Studien und Leitlinien über Migräne in der Schwangerschaft.
UNASSIGNED: Während der Schwangerschaft erfahren bis zu drei Viertel der Frauen eine Verbesserung ihrer Migräne. Allerdings kann es nach der Geburt zu einer erneuten Frequenzzunahme kommen. Die Auswahl einer angemessenen Therapie erfordert eine sorgfältige Nutzen-Risiko-Abwägung. Es ist wichtig, sekundäre Kopfschmerzursachen zu erwägen, da diese in der Schwangerschaft vermehrt auftreten und lebensbedrohlich sein können.
UNASSIGNED: Die Berücksichtigung spezifischer Aspekte der Migräne in der Schwangerschaft ist entscheidend, um die bestmöglichen Behandlungsstrategien für betroffene Patientinnen entwickeln zu können.

BACKGROUND: A previously published web-based App using Gradient-boosted models (GBMs) of eight laboratory parameters was established by Oster et al. to facilitate diagnosis or exclusion of myelodysplastic syndromes (MDS) in patients.
METHODS: To validate their algorithm, we compared 175 anemic patients with MDS diagnosis from our German MDS Registry with 1378 non-MDS anemic patients who consulted various specialties in the Düsseldorf university hospital.
RESULTS: Based on hemoglobin level, leukocyte and platelet count, mean corpuscular volume, absolute neutrophil count, absolute monocyte count, glucose and creatinine, plus the patients\' gender and age, we could not reproduce a high negative predictive value (NPV), but confirmed a useful specificity of 90.9% and a positive predictive value (PPV) of 77.1%. 1192 of 1378 controls were correctly categorized as \"probably not MDS (pnMDS)\" patients. A total of 65 patients were wrongly classified as \"probable MDS (pMDS),\" of whom 48 had alternative explanations for their altered laboratory results. In a second analysis, we included 29 patients with chronic myelomonocytic leukemia (CMML) resulting in only one label as possible MDS, suggesting that highly proliferative bone marrow disorders are correctly excluded.
CONCLUSIONS: The possibility of reliably excluding MDS from differential diagnosis based on peripheral blood lab work appears to be attractive for patients and physicians alike while the confirmation of MDS diagnosis still requires a bone marrow biopsy.