Copenhagen University Hospital
CURIS (Copenhagen University Research Information System)Not a member yet
395145 research outputs found
Sort by
Association between adjuvant chemotherapy and risk of incisional hernia after colorectal cancer resection - a single-centre cohort study
No full textPURPOSE: The study aimed to determine whether adjuvant chemotherapy after colorectal cancer resection is associated with an increased risk of developing incisional hernia (IH).METHODS: All patients undergoing elective curative resection of stage I-III colon adenocarcinoma between 2014 and 2017 at Copenhagen University Hospital - North Zealand were registered in a local database prospectively. This study collected additional data from patients' electronic health records (EHR) on chemotherapy and incision placement. The follow-up computed tomography scans were retrospectively reassessed by two trained surgeons for incisional hernia for up to five years. Patients were stratified into two groups: those receiving adjuvant chemotherapy (study group) and those not receiving it (control group). The primary outcome was CT-verified IH.RESULTS: Of 451 patients, 303 were included in the final analysis, with 108 in the chemotherapy group and 195 in the control group. The study found that the unadjusted risk of IH was 28.9% (95% CI: 20.4-37.4) in the chemotherapy group and 21.1% (95% CI: 15.4-26.9) in the control group with an absolute risk difference of 7.7% (95% CI: -2.3-17.9, p = 0.13). After inverse probability of treatment weighting, the absolute risk difference was 5.6% (95% CI: -4.2-15.6), p = 0.26).CONCLUSIONS: Adjuvant chemotherapy after colon cancer resection does not seem to be associated with a significantly higher risk of IH.TRIAL REGISTRATION: Danish Patient Safety Authority (#31-1522-51).</p
Safety and efficacy assessment of a chewable tablet containing sarolaner, moxidectin, and pyrantel in the treatment of Angiostrongylus vasorum in dogs
No full textAngiostrongylosis, caused by Angiostrongylus vasorum, is an important clinical disease in dogs in Europe with infection transmitted through infected snails and slugs which function as intermediate hosts. Previous studies demonstrated the preventive efficacy of a monthly administration of moxidectin in combination with sarolaner and pyrantel (Simparica Trio® chewable tablet (SMP); Zoetis). An experimental infection study and a field study were conducted to evaluate the efficacy and safety of SMP for the treatment of infections with adult stages of A. vasorum in dogs. The experimental study evaluated the efficacy of a single (n = 7 dogs) or repeated dose (n = 8 dogs) of SMP at the minimum recommended dose of 1.2 mg/kg sarolaner, 0.024 mg/kg moxidectin and 5 mg/kg pyrantel (as pamoate salt) administered as a chewable tablet. against an induced infection with 200 (±10) L3 A. vasorum at least 63 days prior to treatment on Day 0, compared to a placebo control product (n = 8 dogs). Efficacy was determined based on adult worm counts 56 or 57 days after treatment. The field study at veterinary practices in Italy and Denmark evaluated the efficacy and safety of SMP in client-owned dogs (n = 66). Advocate® Spot On (100 mg/mL (minimum 10 mg/kg) imidacloprid + 25 mg/mL (minimum 2.5 mg/kg) moxidectin from Elanco Animal Health, n = 37) was used as the control product. Treatment was administered on Day 0 and repeated on Day 30 if the dog was still positive for A. vasorum larvae. Faecal samples were collected and analyzed with the modified Baermann method in-between scheduled visits. In the experimental study the percentage reduction in geometric mean adult worm counts compared to the placebo-treated group was 98.3 % (P < 0.0001) after a single dose of SMP and 99.6 % (P < 0.0001) after two doses. In the field study, faecal larvae counts were reduced by 90 % or more in 90.9 % of the dogs in the SMP-treated group at Day 30 and in 98.5 % of dogs at Day 60. Compared to the pre-treatment faecal larvae counts a geometric mean percent reduction of 96.7 % was observed on Day 30 and 99.1 % on Day 60 in the SMP treated group. Both in the experimental and in the field study, SMP as a chewable tablet was well tolerated and resulted in a high efficacy against adult A. vasorum infections after a single treatment.</p
Two-stage recurrent events random effects models
No full textWe consider semiparametric random-effects models for recurrent events in the presence of a terminal event. The recurrent events have either a proportional marginal rate model (Cox in J Roy Stat Soc Ser B 34:406–424, 1972) or a proportional marginal mean model (Ghosh and Lin in Stat Sin 34: 663–688, 2002), while the marginal rate of the terminal event is given by a proportional model. The dependency between the recurrent events and the terminal event is described by two variants of random effects models that allow the processes to share the random effect, either fully or partly. The models are formulated as two-stage models, where the marginals can be fitted in an initial stage, and then subsequently random effects parameters can be estimated. The estimation of parameters does not require the choice of any tuning parameters, in contrast to procedures based on numerical integration, and the numerical procedure works well. Standard errors were computed by bootstrapping. The methods are applied to the Taichung Peritoneal Dialysis Study (Chen et al. in Biom J 57(2):215–233, 2015) that considered recurrent inflammations in dialysis patients.We consider semiparametric random-effects models for recurrent events in the presence of a terminal event. The recurrent events have either a proportional marginal rate model (Cox in J Roy Stat Soc Ser B 34:406–424, 1972) or a proportional marginal mean model (Ghosh and Lin in Stat Sin 34: 663–688, 2002), while the marginal rate of the terminal event is given by a proportional model. The dependency between the recurrent events and the terminal event is described by two variants of random effects models that allow the processes to share the random effect, either fully or partly. The models are formulated as two-stage models, where the marginals can be fitted in an initial stage, and then subsequently random effects parameters can be estimated. The estimation of parameters does not require the choice of any tuning parameters, in contrast to procedures based on numerical integration, and the numerical procedure works well. Standard errors were computed by bootstrapping. The methods are applied to the Taichung Peritoneal Dialysis Study (Chen et al. in Biom J 57(2):215–233, 2015) that considered recurrent inflammations in dialysis patients.</p
Microbiota modulation by a human Paneth cell α-defensin fragment protects against DSS-induced colitis
No full textThe gastrointestinal tract hosts a diverse microbial community, named the intestinal microbiota, which plays a vital role in gut health. Host defense peptides (HDPs), such as defensins, help regulate interactions between the host and its microbiota. Dysfunction in HDPs expression has been associated with a greater risk of inflammatory bowel diseases (IBDs), while some HDPs may positively shape the gut environment. Human alpha defensin 5 (HD5) can be cleaved by intestinal enzymes into bioactive fragments such as HD51-9. In this study, we screened various HDPs (LL-37, hBD2, HD5, and HD51-9) for their modulatory effects on healthy human microbiotas in vitro and found distinct, donor-specific effects, with HD51-9 showing the strongest functional impact. Testing HD51-9 in a mouse colitis model revealed potent and microbiota-dependent protection against DSS-induced inflammation, partly through the beneficial of mucus-microbiota interactions.</p
Phylogenomics and the evolution of larval feeding habits in the blow flies (Diptera: Calliphoridae)
No full textBlow flies (Diptera: Calliphoridae) occur worldwide and exhibit a wide range of larval feeding habits, including saprophagy, coprophagy, parasitism and predation. Understanding their biology is critical for medical and veterinary science and ecology. Calliphorids thrive across a range of habitats and exhibit complex life histories, with larvae developing immersed in their food substrate, while adults are free-living and have diverse feeding strategies. Some species have evolved specialized parasitic associations with vertebrate or invertebrate hosts, which are behaviors with important implications for agriculture and for understanding evolutionary transitions between saprophagy and parasitism. This study presents a comprehensive phylogenetic analysis of the Calliphoridae, utilizing 711 of 736 analysed nuclear genes, using anchored hybrid enrichment, from a global collection of blow flies and their relatives. Our results provide a robust and novel reconstruction of the evolutionary history of this group, pinpointing major transitions in larval feeding habits. We argue that saprophagy evolved independently multiple times from invertebrate parasitic ancestors, with vertebrate parasitism emerging from a number of different feeding strategies. These findings challenge prior hypotheses and offer new insights into the adaptive traits driving trophic specialization and diversification in this group.</p
Exposing and Mitigating Calibration Biases and Demographic Unfairness in MLLM Few-Shot In-Context Learning for Medical Image Classification
No full textMultimodal large language models (MLLMs) have enormous potential to perform few-shot in-context learning in the context of medical image analysis. However, safe deployment of these models into real-world clinical practice requires an in-depth analysis of the accuracies of their predictions, and their associated calibration errors, particularly across different demographic subgroups. In this work, we present the first investigation into the calibration biases and demographic unfairness of MLLMs’ predictions and confidence scores in few-shot in-context learning for medical image classification. We introduce CALIN, an inference-time calibration method designed to mitigate the associated biases. Specifically, CALIN estimates the amount of calibration needed, represented by calibration matrices, using a bi-level procedure: progressing from the population level to the subgroup level prior to inference. It then applies this estimation to calibrate the predicted confidence scores during inference. Experimental results on three medical imaging datasets: PAPILA for fundus image classification, HAM10000 for skin cancer classification, and MIMIC-CXR for chest X-ray classification demonstrate CALIN’s effectiveness at ensuring fair confidence calibration in its prediction, while improving its overall prediction accuracies and exhibiting minimum fairness-utility trade-off. The codebase can be found at https://github.com/xingbpshen/medical-calibration-fairness-mllm.Multimodal large language models (MLLMs) have enormous potential to perform few-shot in-context learning in the context of medical image analysis. However, safe deployment of these models into real-world clinical practice requires an in-depth analysis of the accuracies of their predictions, and their associated calibration errors, particularly across different demographic subgroups. In this work, we present the first investigation into the calibration biases and demographic unfairness of MLLMs’ predictions and confidence scores in few-shot in-context learning for medical image classification. We introduce CALIN, an inference-time calibration method designed to mitigate the associated biases. Specifically, CALIN estimates the amount of calibration needed, represented by calibration matrices, using a bi-level procedure: progressing from the population level to the subgroup level prior to inference. It then applies this estimation to calibrate the predicted confidence scores during inference. Experimental results on three medical imaging datasets: PAPILA for fundus image classification, HAM10000 for skin cancer classification, and MIMIC-CXR for chest X-ray classification demonstrate CALIN’s effectiveness at ensuring fair confidence calibration in its prediction, while improving its overall prediction accuracies and exhibiting minimum fairness-utility trade-off. The codebase can be found at https://github.com/xingbpshen/medical-calibration-fairness-mllm.</p
High meiofaunal diversity in the Azores revealed through DNA metabarcoding
No full textTo advance understanding of meiofauna, key components of marine benthic ecosystems, an international team conducted a morphology-based taxonomic survey and workshop around S & atilde;o Miguel, the largest Azorean island, in 2019. The survey yielded critical baseline data on meiofaunal diversity and, despite taxonomic challenges that limited identification of certain taxa, laid a solid foundation for further research. To expand these findings, additional samples were analyzed using DNA metabarcoding, a powerful tool that assesses multiple facets of biodiversity, including richness, community composition, and phylogenetic diversity. Metabarcoding identified 480 meiofaunal amplicon sequence variants across 14 phyla, more than doubling the 180 taxa recorded through morphology alone. Of these, 298 (approximately 62%) belonged to phyla targeted during the morphology-based workshop. A comparison between the two approaches, focusing on clades addressed during the workshop, revealed overlapping but expanded patterns of diversity relative to morphology-based taxonomy. This consistency supports the metabarcoding's ability to uncover meiofaunal diversity and underscores the importance of integrating both methods to achieve a more comprehensive understanding of biodiversity and ecological patterns. As the first metabarcoding study in the Azores, this work establishes a biodiversity baseline for a remote and understudied region, indicating greater benthic diversity than previously recognized. By revealing the complexity of these understudied ecosystems, this research contributes to document and support conservation of biodiversity in the Azores and emphasizes the need for further exploration of meiofaunal communities in isolated oceanic environments