Medical Literature Research Articles

Transformer-Based Language Models for Group Randomized Trial Classification in Biomedical Literature: Model Development and Validation.

For the public health community, monitoring recently published articles is crucial for staying informed about the latest research developments. However, identifying publications about studies with specific research designs from the extensive body of public health publications is a challenge with the currently available methods. Our objective is to develop a fine-tuned pretrained language model that can accurately identify publications from clinical trials that use a group- or cluster-randomized trial (GRT), individually randomized group-treatment trial (IRGT), or stepped wedge group- or cluster-randomized trial (SWGRT) design within the biomedical literature. We fine-tuned the BioMedBERT language model using a dataset of biomedical literature from the Office of Disease Prevention at the National Institute of Health. The model was trained to classify publications into three categories of clinical trials that use nested designs. The model performance was evaluated on unseen data and demonstrated high sensitivity and specificity for each class. When our proposed model was tested for generalizability with unseen data, it delivered high sensitivity and specificity for each class as follows: negatives (0.95 and 0.93), GRTs (0.94 and 0.90), IRGTs (0.81 and 0.97), and SWGRTs (0.96 and 0.99), respectively. Our work demonstrates the potential of fine-tuned, domain-specific language models to accurately identify publications reporting on complex and specialized study designs, addressing a critical need in the public health research community. This model offers a valuable tool for the public health community to directly identify publications from clinical trials that use one of the three classes of nested designs.

Surgical and oncological outcomes in transverse colon carcinoma: does tumor sublocation make a difference?

BackgroundTransverse colon carcinomas are often treated as a single entity in medical literature, despite differences in embryology, anatomy, physiology, genetics, and surgical treatment. We hypothesized that tumor sublocation affects demographics, oncological, surgical and quality of life outcomes.MethodsA retrospective analysis of patients who underwent surgery for transverse colon carcinomas between 2000 and 2018 was conducted. Tumor localization was determined by operative, pathological and imaging reports, and procedures were defined by the extent of vascular resection.ResultsThe study included 273 patients aged 69 years (SD 12.3), of whom 44% were female. The BMI was 28.8 kg/m2 (SD 6.2), and 61% were ASA class 3. Carcinomas were in the proximal (22%), the mid (42%), and the distal transverse colon (36%). Mid-transverse carcinomas exhibited the highest prevalence among female patients (53% vs. 35% proximal vs. 39% distal, p < 0.03). Proximal transverse carcinomas presented with a higher proportion of pathological stage II than mid and distal transverse carcinomas (68.3% vs. proximal vs. 44.3% mid-vs. 49% distal, p = 0.006). On multivariate analysis, anemia was more likely in mid than proximal (p = 0.009) and distal (p = 0.002) transverse colon cancers. Obstruction occurred more often in proximal than mid (p = 0.003), and hematochezia in distal than in mid (p < 0.001) and proximal (p = 0.014) transverse colon carcinomas. The 30-day mortality and morbidity and 5-year overall and disease-free survival rates were similar between the tumor sublocations.ConclusionSublocations of the transverse colon carcinomas affect the symptoms of patients. Tumor sublocation does not impact the intraoperative, postoperative, or oncological outcomes and quality of life.

Efficacy and safety of cinobufacin-based combination therapy for advanced hepatocellular carcinoma: a meta-analysis of randomized controlled trials

BackgroundPrimary liver cancer is one of the most common malignant tumors with a rising incidence in recent years. In China, it ranks as the fourth most prevalent malignancy. Although surgery is the primary treatment, many advanced-stage patients are ineligible due to late diagnosis, rapid progression, or other contraindications. Targeted therapy, chemotherapy, and transarterial chemoembolization (TACE) are common non-surgical approaches, but these often result in significant side effects such as gastrointestinal reactions, bone marrow suppression, and high recurrence rates. In this context, integrative treatment with traditional Chinese medicine (TCM) has shown promise. Cinobufacin, a TCM drug developed in China, has demonstrated superior efficacy and safety when combined with Western treatments in patients with advanced hepatocellular carcinoma (HCC).MethodsWe conducted a systematic review and meta-analysis by searching PubMed, Embase, Cochrane Library, CNKI, WanFang Data, VIP, and the Chinese Biomedical Literature Database (CBM) for randomized controlled trials (RCTs) comparing cinobufacin combined with conventional Western treatments to Western treatments alone in advanced HCC patients. The search covered all databases up to December 30, 2024. Statistical analyses were performed using Stata 15.0, and Review Manager.ResultsA total of 30 studies were included in the meta-analysis. Cinobufacin combination therapy significantly improved the disease control rate (DCR, OR = 2.49, 95% CI 2.01 to 3.07), reduced alpha-fetoprotein (AFP) levels (SMD = − 1.86, 95% CI − 2.58 to − 1.13), alanine aminotransferase (ALT) levels (SMD = − 1.66, 95% CI − 2.48 to − 0.83), and total bilirubin (TBIL) levels (SMD = − 1.82, 95% CI − 2.36 to − 1.28). It also increased white blood cell (WBC) counts (SMD = 0.58, 95% CI 0.06 to 1.11) and improved Karnofsky Performance Status (KPS) scores (SMD = 1.24, 95% CI 0.93 to 1.56). Funnel plots and sensitivity analyses confirmed the robustness of the results.ConclusionCinobufacin combined with Western treatments significantly improves clinical efficacy, reduces adverse drug reactions, and enhances the quality of life for patients with advanced HCC.

GPP 2022: perspectives, sentiments, and feedback from the publications community

Objective To identify the impact and integration of GPP 2022 guidelines on the work processes of publications developed in pharmaceutical, publishing, agency, and academic circles. Methods A cross-sectional survey was conducted for 3 weeks from March 1–22, 2024 among medical publication professionals and healthcare researchers. The survey included questions that have been traditionally challenging in publication workstreams and were disseminated through major medical communications networking groups. The survey results were summarized as a percentage for each answer rounded to one decimal place. Results A total of 100 participants responded to the survey among whom 67.0% were familiar with the GPP 2022 guidelines. To mitigate potential misinterpretation risks, we considered only respondents familiar with the guidelines. More than half of the respondents (55.7%) felt that GPP 2022 was effective in addressing ethical considerations; 34.4% respondents agreed that GPP 2022 provided useful guidance on best practices in transparency and data-sharing. Most survey respondents (54.5%) agreed that the authorship and contributorship criteria were clear in GPP 2022. 37.3% of the survey respondents were familiar with publication extenders and felt that extenders added value to the dissemination of research findings; 39.2% were aware of plain-language summaries and 82.4% viewed them as vital for widespread research accessibility. Conclusion The survey suggests a need to improve awareness and support for implementing transparency initiatives that aid in the publication process. Furthermore, although most respondents were aware of publication extenders, there was a need for clarification and harmonization of guidelines on publication extenders.

How many zebras are there now? An updated report on publications on rare diseases in medical literature

How many zebras are there now? An updated report on publications on rare diseases in medical literature

Performance Comparison of Large Language Models for Efficient Literature Screening

Background: Systematic reviewers face a growing body of biomedical literature, making early-stage article screening increasingly time-consuming. In this study, we assessed six large language models (LLMs)—OpenHermes, Flan T5, GPT-2, Claude 3 Haiku, GPT-3.5 Turbo, and GPT-4o—for their ability to identify randomized controlled trials (RCTs) in datasets of increasing difficulty. Methods: We first retrieved articles from PubMed and used all-mpnet-base-v2 to measure semantic similarity to known target RCTs, stratifying the collection into quartiles of descending relevance. Each LLM then received either verbose or concise prompts to classify articles as “Accepted” or “Rejected”. Results: Claude 3 Haiku, GPT-3.5 Turbo, and GPT-4o consistently achieved high recall, though their precision varied in the quartile with the highest similarity, where false positives increased. By contrast, smaller or older models struggled to balance sensitivity and specificity, with some over-including irrelevant studies or missing key articles. Importantly, multi-stage prompts did not guarantee performance gains for weaker models, whereas single-prompt approaches proved effective for advanced LLMs. Conclusions: These findings underscore that both model capability and prompt design strongly affect classification outcomes, suggesting that newer LLMs, if properly guided, can substantially expedite systematic reviews.

Development and International Application of German Medical Terminology: Structural Features, Translation, and Educational Significance

This article explores the historical evolution, linguistic structure, and international application of medical terminology in the German language. It highlights the significant contributions of German-speaking scholars to the global medical lexicon, particularly in the 19th and 20th centuries, and examines how German's morphological and syntactic features enable precise and complex medical expressions. The paper also addresses challenges in the translation of German medical terms, emphasizing the risk of misinterpretation in cross-linguistic communication. Additionally, it discusses the role of German-language medical literature in shaping international medical education and scientific discourse. By analyzing both linguistic and educational dimensions, the study underscores the continuing relevance of German in the global healthcare and academic landscape.

Rupture of the superficial dorsal vein of the penis

This article presents a clinical case of one of the rarest pathologies in emergency urology, previously described only five times in the global medical literature — rupture of the superficial dorsal vein with penile deviation. Clinical diagnosis of this condition is quite difficult without the use of additional imaging methods, which may not be available in emergency settings. However, establishing an accurate diagnosis in the shortest possible time allows for the most appropriate treatment using conservative or surgical methods. The paper provides an analysis of the causes, differential and clinical diagnosis, management strategies, and therapeutic principles with clear indications for surgical intervention. Currently, there are no clinical guidelines for the management of patients with rupture of the superficial dorsal penile vein. Patient complaints and physical examination, as well as Doppler ultrasound imaging, which allows assessing the vascular network of the penis, including the superficial dorsal vein, can assist in establishing a diagnosis of this condition. Magnetic resonance imaging can help identify tunica albuginea rupture, vascular injury, and penile fracture. If there is any doubt regarding the diagnosis, surgical intervention should be performed, including exploration of the corpora cavernosa and the urethra.

The ‘sero-rationalization’ of the Imperial Japanese military, 1926‒1945

Abstract This article examines how Imperial Japanese military doctors—both Army and Navy medical specialists—employed blood-type analysis in military medicine, from the first military medical publication of blood-type research in 1926 to the end of the Asia-Pacific War in 1945. It explores the military physicians’ quest to investigate the relevance of blood-group knowledge and their attempt to integrate ideas derived from Furukawa Takeji’s Blood Type–Temperament Correlation Theory—the idea that blood type is linked to personality traits—into the operations of the armed forces, a process I term ‘sero-rationalization’. By the mid-1930s, however, escalating conflicts prompted a shift in research priorities. Military physicians increasingly focused on serology and the technological advancements required for blood transfusions, moving away from earlier biopsychological discussions of blood types. This shift reflected an urgent need to address wartime medical challenges, including treating injuries and developing reliable transfusion methods. With the intensification of war by the 1940s, frontline physicians began exploring alternatives to traditional blood typing, such as cross-type transfusions and even animal-to-human transfusions. In their attempts to circumvent the ABO blood-group system in dealing with wartime medical emergencies, military physicians departed significantly from their initial emphasis on serological differentiation. Ironically, the pursuit of sero-rationalization—intended to optimize military efficiency—ultimately proved counterproductive.

Cervical Lymphangiolipoleiomyoma: An Unusual Offender at a Usual Destination – Case Report and Literature Review

ABSTRACT Cervical fibroids are relatively rare compared to fibroids that develop in other parts of the uterus, accounting for nearly 1%–2% of all fibroids. They typically originate from the supravaginal or vaginal portions of the cervix. One histological variant is angiolipoleiomyoma (ALLM), a benign mixed mesenchymal neoplasm composed of smooth muscle, mature adipose tissue, and abnormal blood vessels. While there are case reports of ALLM occurring in the kidneys, its presence in the female genital tract is extremely uncommon. Furthermore, here, we present a novel case of a 40-year-old postmenopausal female who presented with a long history of vaginal bleeding. Ultrasound revealed a fibroid located at the cervix; however, with histopathology revealing it to be a lymphangiolipoleiomyoma with a plethora of lymphatics. To the best of our knowledge, this is the first such case described in the medical literature.

Sefer Asaf: Early Medieval Jewish Theory and Practice from the Middle East to Byzantine Italy

Abstract The origins of Sefer Asaf can be reconstructed as follows. In the mid-eighth century a group of Jews learned medicine from dyophysite Christian monks in Western Iran. They paraphrased a few short Syriac texts in Hebrew and used them as reference books. During the ninth century, some of these Jewish healers migrated to Southern Italy, where they learned Greek and studied Greek medical literature. They paraphrased some works of Hippocrates and a Greek uroscopic compendium in Hebrew. They added these new texts to the earlier corpus translated from Syriac, and at some point before the eleventh century they called this compendium Sefer Asaf.

Artificial Intelligence Performs Well for Patient-Level Education of Benign Anorectal Conditions.

BackgroundChatGPT, the most widely recognized AI chatbot, has been increasingly utilized by patients for health care education, but its performance and knowledge base have not been evaluated for colorectal surgery topics. We hypothesize ChatGPT can provide accurate patient-level information for benign anorectal diseases.MethodsWe performed a single-institution prospective study evaluating OpenAI's GPT-4 chatbot against verified online medical literature using the modified EQIP (mEQIP) for hemorrhoids, anal fissures, and pruritus ani. Scoring was performed by three independent educated reviewers.ResultsChatGPT had a median overall score of 22/36 (61%) across all topics. It performed similarly for all topics: 24 for hemorrhoids, 22 for fissures, and 20 for pruritus (P = 0.15). ChatGPT was strongest within the mEQIP Content domain for all topics (P = 0.001). It performed weakest within the mEQIP Identification domain. It primarily lost points for recommending non-evidence-based treatments, lack of citations and visual aids, and generating broken source links.DiscussionChatGPT can provide accurate, reliable information at patient understanding level and is comparable to other validated online information for benign anorectal pathologies. It could improve on mEQIP performance by improving source documentation and visual aid capability, but it remains a promising patient resource with an intuitive user interface.

Modern clinical and laboratory indicators of severe Plasmodium falciparum malaria

According to WHO Report 2024, malaria is still a global health challenge, especially in Africa, Asia and South America. In the opinion of the authors, errors made in adequate assessment of clinical and laboratory indicators of P. falciparum malaria (tropical malaria) in a patient are the major reasons that can lead to a lethal disease outcome. The aim of our research was to describe the clinical and laboratory indicators of severe tropical malaria with a severe and complicated course of disease in a female patient, taking into account the analysis of data from modern scientific and medical literature, and our own experience. An abstract of the case history of patient M., 56 years old, is presented. The patient underwent medical treatment in an infectious disease hospital under the primary diagnosis of Plasmodium falciparum malaria with cerebral complications and acute renal failure, severe course of illness. She fell ill on Day 7 after her journey to Tanzania (Zanzibar Island). The disease developed in a severe, complicated form with acute renal failure, sepsis, severe anemia, bilateral polysegmental pneumonia, and nosebleeds. The parasitemia level reached 1,428,000 per microliter of blood. The treatment regimen of malaria included Coartem. Due to ongoing intensive therapy, positive changes were achieved. A stable absence of parasitemia was observed, however a peripheral edema and a high level of azotemia persisted. The patient continued to undergo treatment in the Department of Nephrology, where a significant clinical outcome in the form of decreased azotemia levels was achieved. The woman was discharged home in a satisfactory condition. The clinical case example fully reflects modern clinical and laboratory criteria for severe, complicated tropical malaria, and demonstrates rapid development of serious complications as a result of late admission to hospital. It represents a practical interest not only for infectious disease experts, but also for therapists, nephrologists, neurologists, resuscitationists and physicians of other specialties.

The influence of prompt engineering on large language models for protein–protein interaction identification in biomedical literature

Identifying protein–protein interactions (PPIs) is a foundational task in biomedical natural language processing. While specialized models have been developed, the potential of general-domain large language models (LLMs) in PPI extraction, particularly for researchers without computational expertise, remains unexplored. This study evaluates the effectiveness of proprietary LLMs (GPT-3.5, GPT-4, and Google Gemini) in PPI prediction through systematic prompt engineering. We designed six prompting scenarios of increasing complexity, from basic interaction queries to sophisticated entity-tagged formats, and assessed model performance across multiple benchmark datasets (LLL, IEPA, HPRD50, AIMed, BioInfer, and PEDD). Carefully designed prompts effectively guided LLMs in PPI prediction. Gemini 1.5 Pro achieved the highest performance across most datasets, with notable F1-scores in LLL (90.3%), IEPA (68.2%), HPRD50 (67.5%), and PEDD (70.2%). GPT-4 showed competitive performance, particularly in the LLL dataset (87.3%). We identified and addressed a positive prediction bias, demonstrating improved performance after evaluation refinement. While not surpassing specialized models, general-purpose LLMs with appropriate prompting strategies can effectively perform PPI prediction tasks, offering valuable tools for biomedical researchers without extensive computational expertise.

Research Methodology in Acupuncture for Managing Interstitial Cystitis/Bladder Pain Syndrome: A Scoping Review.

Acupuncture and moxibustion are being explored as complementary therapies with promising potential in managing interstitial cystitis/bladder pain syndrome (IC/BPS). While some studies indicate possible benefits, their scientific basis and clinical efficacy remain subjects of debate, partly due to methodological flaws in study designs involving acupuncture and related modalities, which warrant further attention and discussion. The main objective is to analyze and evaluate the research methodologies used in existing literature on acupuncture for managing IC/BPS, providing insight into the methodological challenges and opportunities in this field. National Library of Medicine (PubMed), Allied and Complementary Medicine Database (AMED), Cochrane Library, Excerpta Medica database (EMBASE), Web of Sciences (WOS), Medical Literature Analysis and Retrieval System Online (MEDLINE), Cumulated Index in Nursing and Allied Health Literature (CINAHL), Scopus, SPORTDiscus, Epistemonikos, and Physiotherapy Evidence Database (PEDro) were searched from their inception to August 2024. Data were extracted based on the study designs, primary outcome measures, adverse events (AEs), and participants' subjective views. All the experimental studies (27.3%, 6 of 22) were randomized controlled trials (RCTs), but only 2 full texts were accessible. Of the 4 RCTs, 3 involved rats. Observational research (50%) included case reports (27.3%), cohort studies (9.1%), and theoretical studies (4.5%). Additionally, reviews (22.7%) were included. The main languages were English (77.3%), Chinese (13.6%), Japanese (4.5%), and Russian (4.5%). The most frequently reported outcome was pain, followed by quality-of-life impact (symptom burden), urinary frequency and urgency, nocturia, and bladder capacity. AEs were reported in only 5 articles, all of which concluded that these effects were not significant and that acupuncture could be considered a safe and relatively noninvasive technique. One review found a risk of 1 AE/76 000 patients, with the most common side effects being minor, such as bleeding or bruising. Current evidence on acupuncture for interstitial cystitis is limited in quality, with studies often compromised by biases, small sample sizes, and lack of standardized protocols. Acupuncture is frequently part of multimodal treatments, aligning with Traditional Chinese Medicine's holistic approach to balance physical, mental, and emotional health. More high-quality experimental research is needed, focusing on standardized protocols and participant experiences to better evaluate its efficacy and safety.

Leveraging long context in retrieval augmented language models for medical question answering

While holding great promise for improving and facilitating healthcare through applications of medical literature summarization, large language models (LLMs) struggle to produce up-to-date responses on evolving topics due to outdated knowledge or hallucination. Retrieval-augmented generation (RAG) is a pivotal innovation that improves the accuracy and relevance of LLM responses by integrating LLMs with a search engine and external sources of knowledge. However, the quality of RAG responses can be largely impacted by the rank and density of key information in the retrieval results, such as the “lost-in-the-middle” problem. In this work, we aim to improve the robustness and reliability of the RAG workflow in the medical domain. Specifically, we propose a map-reduce strategy, BriefContext, to combat the “lost-in-the-middle” issue without modifying the model weights. We demonstrated the advantage of the workflow with various LLM backbones and on multiple QA datasets. This method promises to improve the safety and reliability of LLMs deployed in healthcare domains by reducing the risk of misinformation, ensuring critical clinical content is retained in generated responses, and enabling more trustworthy use of LLMs in critical tasks such as medical question answering, clinical decision support, and patient-facing applications.

Navigating Private Equity Buyouts: A Guide for Plastic Surgeons at All Career Stages.

Recently, private equity (PE) investments in plastic surgery have increased as senior surgeons retire and transfer their legacy practices to corporate ownership in exchange for substantial payouts. During this transition, clinicians and practitioners at acquired practices often feel skeptical about the organization's goals and future. If left unaligned, the relationship between these groups deteriorates, and consequently, so does the quality of clinical care. Although the medical literature on this topic highlights the risks of PE, it often fails to address strategies for fostering a healthy working relationship among these groups. A comprehensive review of the medical business management literature was conducted to identify potential solutions applicable to clinical settings. PE acquisitions typically give rise to 5 core concerns: future uncertainty, salary cuts or job losses, increased workloads, loss of clinical autonomy, and unclear organizational and career prospects. This article identifies recommendations tailored to both clinicians and PE leaders to improve integration, alignment, and productive partnerships. These recommendations address the unique concerns and aspirations of physicians at various stages of their careers within the PE landscape, with the overarching goal of enhancing practice valuation and achieving career success without compromising patient care. Drawing insights from business management, this article proposes actionable steps to improve the integration of PE firms into plastic surgery practices. A collaborative partnership can benefit all stakeholders-physicians, practices, PE investors, and patients-while safeguarding the quality of patient care.

SB-AGT: A stochastic beam search-enhanced attention-based Gumbel tree framework for drug-drug interaction extraction from biomedical literature.

SB-AGT: A stochastic beam search-enhanced attention-based Gumbel tree framework for drug-drug interaction extraction from biomedical literature.

Yeast Knowledge Graphs Database for Exploring Saccharomyces Cerevisiae and Schizosaccharomyces Pombe.

Biomedical literature contains an extensive wealth of information on gene and protein function across various biological processes and diseases. However, navigating this vast and often restricted-access data can be challenging, making it difficult to extract specific insights efficiently. In this study, we introduce a high-throughput pipeline that leverages OpenAI's Generative Pre-Trained Transformer Model (GPT) to automate the extraction and analysis of gene function information. We applied this approach to 84,427 publications on Saccharomyces cerevisiae and 6,452 publications on Schizosaccharomyces pombe, identifying 3,432,749 relationships for budding yeast and 421,198 relationships for S. pombe. This resulted in a comprehensive, searchable online Knowledge Graph database, available at yeast.connectome.tools and spombe.connectome.tools, which offers users extensive access to various interactions and pathways. Our analysis underscores the power of integrating artificial intelligence with bioinformatics, as demonstrated through key insights into important nodes like Hsp104 and Atg8 proteins. This work not only facilitates efficient data extraction in yeast research but also presents a scalable model for similar studies in other biological systems.

Cesarean section with operation on placental support as an adapted EXIT procedure for fetal thyroid goiter: A case report and literature review.

The ex-utero intrapartum treatment (EXIT) procedure is a specialized delivery strategy that extends utero-placental-fetal circulation to convert a potential neonatal emergency condition into a condition that is compatible with postnatal life. Cesarean section with operation on placental support is an EXIT technique that requires a relatively short duration of placental support and few skilled medical personnel and specialized instruments; it can successfully treat selected fetal indications. In the present study, we report a case of fetal thyroid goiter as an example of a fetal anomaly requiring the procedure. We then review all cases published in the medical literature that were similar to our procedure (15 cases including our new case). Twelve (80%) cases of fetal anomalies and three prophylactic cases of extremely low birth weight were delivered using a procedure adapted from the standard EXIT procedure. All 12 cases of fetal anomalies were treated by airway intervention. In 12 of the 15 cases (80%), direct laryngoscopy and endotracheal intubation were attempted; intubation was successful in seven of these 12 cases (58%). There was only one case of postpartum hemorrhage caused by uterine incisional bleeding, with a consequence of delayed hysterotomy closure. In summary, cesarean section with operation on placental support is a promising alternative delivery technique for neonates with airway obstructive lesions, especially in resource-limited settings.