Search

Wir Menschen sind auf funktionierende und widerstandsfähige Ökosysteme angewiesen. Der Zustand der Natur in Deutschland hat sich in den letzten Jahrzehnten weiter verschlechtert. Ergänzend zum Schutz der verbliebenen Natur sollte stärker als bisher der Zustand geschädigter Ökosysteme verbessert werden. SRU, WBBGR und WBW geben mit dieser Stellungnahme Empfehlungen für eine wirksame Renaturierungspolitik in Deutschland. Am 23. Mai 2024 wurden die Kernempfehlungen der Stellungnahme in einer öffentlichen Online-Veranstaltung präsentiert und diskutiert. Die Dokumentation ist hier abrufbar. Das Anliegen, die Erholung eines degradierten Ökosystems zu unterstützen, ist weit zu verstehen und umfasst neben der Verbesserung des Zustands von Schutzgebieten auch naturverträgliche Formen der Landnutzung. Ökosysteme sollen sich durch Renaturierung in Richtung naturnäherer Strukturen entwickeln, sodass sie langfristig vielfältige Leistungen erbringen können. Wie andere Vorhaben benötigen Renaturierungsmaßnahmen Flächen. Ökosysteme wiederherzustellen wird daher auch Konflikte um Flächennutzung, wirtschaftliche Aspekte und unser Konsumverhalten mit sich bringen. Die es zu moderieren und zu entscheiden gilt. Renaturierung ist eine gesellschaftliche Aufgabe, die nicht nur den Naturschutz betrifft, sondern vor allem auch die landnutzenden Sektoren, neben der Stadt- und Regionalplanung insbesondere die Landwirtschaft und die Waldbewirtschaftung. Daher zielt diese Stellungnahme darauf ab, Grundzüge einer Renaturierungspolitik zu entwickeln, die Synergien zwischen Naturschutz und Landnutzungsinteressen erzeugt sowie Zielkonflikte minimiert.

Little is known about the strong mediating effect of the ligand sphere and the coordination geometry on the strength and isotopologue selectivity of hydrogen adsorption on the undercoordinated copper(I) site. Here, we explore this effect using gas-phase complexes Cu+(H2O)(H2)n (with n ≤ 3) as model systems. Cu+(H2O) attracts dihydrogen (82 kJ mol−1) more strongly than bare Cu+ (64 kJ mol−1) does. Combining experimental and computational methods, we demonstrate a high isotopologue selectivity in dihydrogen binding to Cu+(H2O), which results from a large difference in the adsorption zero-point energies (2.8 kJ mol−1 between D2 and H2, including an anharmonic contribution of 0.4 kJ mol−1). We investigate its origins and the bond strengthening between Cu+ and H2 upon addition of a single H2O ligand. We discuss the role of the environment and the coordination geometry of the adsorption site in achieving a high selectivity and the ramifications for identifying and designing future materials for adsorptive dihydrogen isotopologue separation.

The dataset consists of Infrared photodissociation (IRPD) spectra of Cu+(H2O)(H2)n (with n ≤ 3) and its isotopologue measured on the Leipzig 5 K ring-electrode ion-trap triple mass spectrometer. Besides, it contains the Energy Decomposition Analysis (EDA), the benchmark results, the harmonic and the anharmonic VPT2 frequencies results as well as the script used to get the predicted separation factor for the adsorbed dihydrogen isotopologue.

Background: The internet is a common source of health information for patients and caregivers. To date, content and information quality of YouTube videos on sarcoidosis has not been studied. The aim of our study was to investigate the content and quality of information on sarcoidosis provided by YouTube videos. Methods: Of the first 200 results under the search term “sarcoidosis,” all English-language videos with content directed at patients were included. Two independent investigators assessed the content of the videos based on 25 predefined key features (content score with 0–25 points), as well as reliability and quality (HONCode score with 0–8 points, DISCERN score with 1–5 points). Misinformation contained in the videos was described qualitatively. Results: The majority of the 85 included videos were from an academic or governmental source (n = 63, 74%), and median time since upload was 33 months (IQR 10–55). Median video duration was 8 min (IQR 3–13) and had a median of 2,044 views (IQR 504 − 13,203). Quality assessment suggested partially sufficient information: mean HONCode score was 4.4 (SD 0.9) with 91% of videos having a medium quality HONCode evaluation. Mean DISCERN score was 2.3 (SD 0.5). Video content was generally poor with a mean of 10.5 points (SD 0.6). Frequently absent key features included information on the course of disease (6%), presence of substantial geographical variation (7%), and importance of screening for extrapulmonary manifestations (11%). HONCode scores were higher in videos from academic or governmental sources (p = 0.003), particularly regarding “transparency of sponsorship” (p < 0.001). DISCERN and content scores did not differ by video category. Conclusions: Most YouTube videos present incomplete information reflected in a poor content score, especially regarding screening for extrapulmonary manifestations. Quality was partially sufficient with higher scores in videos from academic or governmental sources, but often missing

Background: Fibrotic interstitial lung diseases (fILDs) are a heterogeneous group of lung diseases associated with significant morbidity and mortality. Despite a large increase in the number of clinical trials in the last 10 years, current regulatory-approved management approaches are limited to two therapies that prevent the progression of fibrosis. The drug development pipeline is long and there is an urgent need to accelerate this process. This manuscript introduces the concept and design of an innovative research approach to drug development in fILD: a global Randomised Embedded Multifactorial Adaptive Platform in fILD (REMAP-ILD). Methods: Description of the REMAP-ILD concept and design: the specific terminology, design characteristics (multifactorial, adaptive features, statistical approach), target population, interventions, outcomes, mission and values, and organisational structure. Results: The target population will be adult patients with fILD, and the primary outcome will be a disease progression model incorporating forced vital capacity and mortality over 12 months. Responsive adaptive randomisation, prespecified thresholds for success and futility will be used to assess the effectiveness and safety of interventions. REMAP-ILD embraces the core values of diversity, equity, and inclusion for patients and researchers, and prioritises an open-science approach to data sharing and dissemination of results. Conclusion: By using an innovative and efficient adaptive multi-interventional trial platform design, we aim to accelerate and improve care for patients with fILD. Through worldwide collaboration, novel analytical methodology and pragmatic trial delivery, REMAP-ILD aims to overcome major limitations associated with conventional randomised controlled trial approaches to rapidly improve the care of people living with fILD.

Background:There is no standard definition of respiratory-related hospitalisation, a common end-point in idiopathic pulmonary fibrosis (IPF) clinical trials. As diverse aetiologies and complicating comorbidities can present similarly, external adjudication is sometimes employed to achieve standardisation of these events. Methods:An algorithm for respiratory-related hospitalisation was developed through a literature review of IPF clinical trials with respiratory-related hospitalisation as an end-point. Experts reviewed the algorithm until a consensus was reached. The algorithm was validated using data from the phase 3 ISABELA trials (clinicaltrials.gov identifiers NCT03711162 and NCT03733444), by assessing concordance between nonadjudicated, investigator-defined, respiratory-related hospitalisations and those defined by the adjudication committee using the algorithm. Results:The algorithm classifies respiratory-related hospitalisation according to cause: extraparenchymal (worsening respiratory symptoms due to left heart failure, volume overload, pulmonary embolism, pneumothorax or trauma); other (respiratory tract infection, right heart failure or exacerbation of COPD); “definite” acute exacerbation of IPF (AEIPF) (worsening respiratory symptoms within 1 month, with radiological or histological evidence of diffuse alveolar damage); or “suspected” AEIPF (as for “definite” AEIPF, but with no radiological or histological evidence of diffuse alveolar damage). Exacerbations (“definite” or “suspected”) with identified triggers (infective, post-procedural or traumatic, drug toxicity-or aspiration-related) are classed as “known AEIPF”; “idiopathic AEIPF” refers to exacerbations with no identified trigger. In the ISABELA programme, there was 94% concordance between investigator-and adjudication committee-determined causes of respiratory-related hospitalisation. Conclusion:The algorithm could help

BACKGROUND Fibrotic interstitial lung diseases (fILDs) are a heterogeneous group of lung diseases associated with significant morbidity and mortality. Despite a large increase in the number of clinical trials in the last 10 years, current regulatory-approved management approaches are limited to two therapies that prevent the progression of fibrosis. The drug development pipeline is long and there is an urgent need to accelerate this process. This manuscript introduces the concept and design of an innovative research approach to drug development in fILD: a global Randomised Embedded Multifactorial Adaptive Platform in fILD (REMAP-ILD). METHODS Description of the REMAP-ILD concept and design: the specific terminology, design characteristics (multifactorial, adaptive features, statistical approach), target population, interventions, outcomes, mission and values, and organisational structure. RESULTS The target population will be adult patients with fILD, and the primary outcome will be a disease progression model incorporating forced vital capacity and mortality over 12 months. Responsive adaptive randomisation, prespecified thresholds for success and futility will be used to assess the effectiveness and safety of interventions. REMAP-ILD embraces the core values of diversity, equity, and inclusion for patients and researchers, and prioritises an open-science approach to data sharing and dissemination of results. CONCLUSION By using an innovative and efficient adaptive multi-interventional trial platform design, we aim to accelerate and improve care for patients with fILD. Through worldwide collaboration, novel analytical methodology and pragmatic trial delivery, REMAP-ILD aims to overcome major limitations associated with conventional randomised controlled trial approaches to rapidly improve the care of people living with fILD.

This article summarises a selection of scientific highlights in the field of interstitial lung diseases (ILDs) presented at the International Congress of the European Respiratory Society in 2023. Translational and clinical studies focused on the whole spectrum of ILDs, from (ultra)rare ILDs to sarcoidosis, ILDs associated with connective tissue disease and idiopathic pulmonary fibrosis. The main topics of the 2023 Congress presentations were improving the diagnostic process of ILDs, better prediction of disease course and investigation of novel treatment options.

BACKGROUND: Dyspnoea and cough can have a profound impact on the lives of patients with pulmonary fibrosis. We investigated the effects of nintedanib on the symptoms and impact of pulmonary fibrosis in patients with progressive pulmonary fibrosis (PPF) in the INBUILD trial using the Living with Pulmonary Fibrosis (L-PF) questionnaire. METHODS: Patients had a fibrosing interstitial lung disease (ILD) (other than idiopathic pulmonary fibrosis) of >10% extent on high-resolution computed tomography (HRCT) and met criteria for ILD progression within the prior 24 months. Patients were randomised 1:1 to receive nintedanib or placebo. Changes in L-PF questionnaire scores from baseline to week 52 were assessed using mixed models for repeated measures. RESULTS:In total, 663 patients were treated. Compared with placebo, there were significantly smaller increases (worsenings) in adjusted mean L-PF questionnaire total (0.5 versus 5.1), symptoms (1.3 versus 5.3), dyspnoea (4.3 versus 7.8) and fatigue (0.7 versus 4.0) scores in the nintedanib group at week 52. L-PF questionnaire cough score decreased in the nintedanib group and increased in the placebo group (-1.8 versus 4.3). L-PF questionnaire impacts score decreased slightly in the nintedanib group and increased in the placebo group (-0.2 versus 4.6). Similar findings were observed in patients with a usual interstitial pneumonia-like fibrotic pattern on HRCT and in patients with other fibrotic patterns on HRCT.CONCLUSION: Based on changes in L-PF questionnaire scores, nintedanib reduced worsening of dyspnoea, fatigue and cough and the impacts of ILD over 52 weeks in patients with PPF.

Whether future AI models make the world safer or less safe for humans rests in part on our ability to efficiently collect accurate data from people about what they want the models to do. However, collecting high quality data is difficult, and most AI/ML researchers are not trained in data collection methods. The growing emphasis on data-centric AI highlights the potential of data to enhance model performance. It also reveals an opportunity to gain insights from survey methodology, the science of collecting high-quality survey data. In this position paper, we summarize lessons from the survey methodology literature and discuss how they can improve the quality of training and feedback data, which in turn improve model performance. Based on the cognitive response process model, we formulate specific hypotheses about the aspects of label collection that may impact training data quality. We also suggest collaborative research ideas into how possible biases in data collection can be mitigated, making models more accurate and human-centric., Comment: 9 pages, 4 figures. Position paper, under review

Despite the predominance of English in their training data, English-centric Large Language Models (LLMs) like GPT-3 and LLaMA display a remarkable ability to perform multilingual tasks, raising questions about the depth and nature of their cross-lingual capabilities. This paper introduces the decomposed prompting approach to probe the linguistic structure understanding of these LLMs in sequence labeling tasks. Diverging from the single text-to-text prompt, our method generates for each token of the input sentence an individual prompt which asks for its linguistic label. We assess our method on the Universal Dependencies part-of-speech tagging dataset for 38 languages, utilizing both English-centric and multilingual LLMs. Our findings show that decomposed prompting surpasses the iterative prompting baseline in efficacy and efficiency under zero- and few-shot settings. Further analysis reveals the influence of evaluation methods and the use of instructions in prompts. Our multilingual investigation shows that English-centric language models perform better on average than multilingual models. Our study offers insights into the multilingual transferability of English-centric LLMs, contributing to the understanding of their multilingual linguistic knowledge., Comment: 18 pages, 7 figures

The open-ended nature of language generation makes the evaluation of autoregressive large language models (LLMs) challenging. One common evaluation approach uses multiple-choice questions (MCQ) to limit the response space. The model is then evaluated by ranking the candidate answers by the log probability of the first token prediction. However, first-tokens may not consistently reflect the final response output, due to model's diverse response styles such as starting with "Sure" or refusing to answer. Consequently, MCQ evaluation is not indicative of model behaviour when interacting with users. But by how much? We evaluate how aligned first-token evaluation is with the text output along several dimensions, namely final option choice, refusal rate, choice distribution and robustness under prompt perturbation. Our results show that the two approaches are severely misaligned on all dimensions, reaching mismatch rates over 60%. Models heavily fine-tuned on conversational or safety data are especially impacted. Crucially, models remain misaligned even when we increasingly constrain prompts, i.e., force them to start with an option letter or example template. Our findings i) underscore the importance of inspecting the text output, too and ii) caution against relying solely on first-token evaluation.

Prompt-based methods have been successfully applied to multilingual pretrained language models for zero-shot cross-lingual understanding. However, most previous studies primarily focused on sentence-level classification tasks, and only a few considered token-level labeling tasks such as Named Entity Recognition (NER) and Part-of-Speech (POS) tagging. In this paper, we propose Token-Level Prompt Decomposition (ToPro), which facilitates the prompt-based method for token-level sequence labeling tasks. The ToPro method decomposes an input sentence into single tokens and applies one prompt template to each token. Our experiments on multilingual NER and POS tagging datasets demonstrate that ToPro-based fine-tuning outperforms Vanilla fine-tuning and Prompt-Tuning in zero-shot cross-lingual transfer, especially for languages that are typologically different from the source language English. Our method also attains state-of-the-art performance when employed with the mT5 model. Besides, our exploratory study in multilingual large language models shows that ToPro performs much better than the current in-context learning method. Overall, the performance improvements show that ToPro could potentially serve as a novel and simple benchmarking method for sequence labeling tasks., Comment: EACL 2024

OBJECTIVE: To investigate the rate of decline in forced vital capacity (FVC), and the effect of nintedanib on the rate of decline in FVC, in subjects with systemic sclerosis-associated interstitial lung disease (SSc-ILD) who had risk factors for rapid decline in FVC. METHODS: The SENSCIS trial enrolled subjects with SSc and fibrotic ILD of ≥10% extent on high-resolution CT. The rate of decline in FVC over 52 weeks was analysed in all subjects and in those with early SSc (<18 months since first non-Raynaud symptom), elevated inflammatory markers (C reactive protein ≥6 mg/L and/or platelets ≥330×109/L) or significant skin fibrosis (modified Rodnan skin score (mRSS) 15-40 or mRSS ≥18) at baseline. RESULTS: In the placebo group, the rate of decline in FVC was numerically greater in subjects with <18 months since first non-Raynaud symptom (-167.8 mL/year), elevated inflammatory markers (-100.7 mL/year), mRSS 15-40 (-121.7 mL/year) or mRSS ≥18 (-131.7 mL/year) than in all subjects (-93.3 mL/year). Nintedanib reduced the rate of FVC decline across subgroups, with a numerically greater effect in patients with these risk factors for rapid FVC decline. CONCLUSION: In the SENSCIS trial, subjects with SSc-ILD who had early SSc, elevated inflammatory markers or extensive skin fibrosis had a more rapid decline in FVC over 52 weeks than the overall trial population. Nintedanib had a numerically greater effect in patients with these risk factors for rapid ILD progression.

The new European General Data Protection Regulation (GDPR) imposes enhanced requirements on digital data collection. This article reports from a 2018 German nationwide population-based probability app study in which participants were asked through a GDPR compliant consent process to share a series of digital trace data, including geolocation, accelerometer data, phone and text messaging logs, app usage, and access to their address books. With about 4,300 invitees and about 650 participants, we demonstrate (1) people were just as willing to share such extensive digital trace data as they were in studies with far more limited requests; (2) despite being provided more decision-related information, participants hardly differentiated between the different data requests made; and (3) once participants gave consent, they did not tend to revoke it. We also show (4) evidence for a widely-held belief that explanations regarding data collection and data usage are often not read carefully, at least not within the app itself, indicating the need for research and user experience improvement to adequately inform and protect participants. We close with suggestions to the field for creating a seal of approval from professional organizations to help the research community promote the safe use of data.

Introduction There is an unmet need for new treatments for idiopathic pulmonary fibrosis (IPF). The oral preferential phosphodiesterase 4B inhibitor, BI 1015550, prevented a decline in forced vital capacity (FVC) in a phase II study in patients with IPF. This study design describes the subsequent pivotal phase III study of BI 1015550 in patients with IPF (FIBRONEER-IPF).Methods and analysis In this placebo-controlled, double-blind, phase III trial, patients are being randomised in a 1:1:1 ratio to receive 9 mg or 18 mg of BI 1015550 or placebo two times per day over at least 52 weeks, stratified by use of background antifibrotics (nintedanib/pirfenidone vs neither). The primary endpoint is the absolute change in FVC at week 52. The key secondary endpoint is a composite of time to first acute IPF exacerbation, hospitalisation due to respiratory cause or death over the duration of the trial.Ethics and dissemination The trial is being carried out in compliance with the ethical principles of the Declaration of Helsinki, in accordance with the International Council on Harmonisation Guideline for Good Clinical Practice and other local ethics committees. The results of the study will be disseminated at scientific congresses and in peer-reviewed publications.

There is a major need for effective, well-tolerated treatments for idiopathic pulmonary fibrosis (IPF)., info:eu-repo/semantics/published

INTRODUCTION: Progressive pulmonary fibrosis (PPF) includes any diagnosis of progressive fibrotic interstitial lung disease (ILD) other than idiopathic pulmonary fibrosis (IPF). However, disease progression appears comparable between PPF and IPF, suggesting a similar underlying pathology relating to pulmonary fibrosis. Following positive results in a phase II study in IPF, this phase III study will investigate the efficacy and safety of BI 1015550 in patients with PPF (FIBRONEER-ILD). METHODS AND ANALYSIS: In this phase III, double-blind, placebo-controlled trial, patients are being randomised 1:1:1 to receive BI 1015550 (9 mg or 18 mg) or placebo twice daily over at least 52 weeks, stratified by background nintedanib use. Patients must be diagnosed with pulmonary fibrosis other than IPF that is progressive, based on predefined criteria. Patients must have forced vital capacity (FVC) ≥45% predicted and haemoglobin-corrected diffusing capacity of the lung for carbon monoxide ≥25% predicted. Patients must be receiving nintedanib for at least 12 weeks, or not receiving nintedanib for at least 8 weeks, prior to screening. Patients on stable treatment with permitted immunosuppressives (eg, methotrexate, azathioprine) may continue their treatment throughout the trial. Patients with clinically significant airway obstruction or other pulmonary abnormalities, and those using immunosuppressives that may confound FVC results (cyclophosphamide, tocilizumab, mycophenolate, rituximab) or high-dose steroids will be excluded. The primary endpoint is absolute change from baseline in FVC (mL) at week 52. The key secondary endpoint is time to the first occurrence of any acute ILD exacerbation, hospitalisation for respiratory cause or death, over the duration of the trial. ETHICS AND DISSEMINATION: The trial is being carried out in accordance with the ethical principles of the Declaration of Helsinki, the International Council on Harmonisation Guideline for Good Clinical Practice an

IntroductionThere is an unmet need for new treatments for idiopathic pulmonary fibrosis (IPF). The oral preferential phosphodiesterase 4B inhibitor, BI 1015550, prevented a decline in forced vital capacity (FVC) in a phase II study in patients with IPF. This study design describes the subsequent pivotal phase III study of BI 1015550 in patients with IPF (FIBRONEER-IPF). METHODS AND ANALYSIS: In this placebo-controlled, double-blind, phase III trial, patients are being randomised in a 1:1:1 ratio to receive 9 mg or 18 mg of BI 1015550 or placebo two times per day over at least 52 weeks, stratified by use of background antifibrotics (nintedanib/pirfenidone vs neither). The primary endpoint is the absolute change in FVC at week 52. The key secondary endpoint is a composite of time to first acute IPF exacerbation, hospitalisation due to respiratory cause or death over the duration of the trial. ETHICS AND DISSEMINATION: The trial is being carried out in compliance with the ethical principles of the Declaration of Helsinki, in accordance with the International Council on Harmonisation Guideline for Good Clinical Practice and other local ethics committees. The results of the study will be disseminated at scientific congresses and in peer-reviewed publications. TRIAL REGISTRATION NUMBER: NCT05321069.

Introduction: Many patients use the internet as a source of health information. Sarcoidosis is a complex disease, and internet resources have not yet been analyzed for reliability and content on sarcoidosis. Aims: Our study aimed to investigate the content and the quality of information on sarcoidosis provided by internet resources. Methods: Google, Yahoo, and Bing were searched for the term “sarcoidosis,” and the first 200 hits were saved in each case. Those websites that met the inclusion criteria (English language, no registration fees, and relevant to sarcoidosis) were then analyzed by two independent investigators for readability, quality (HON, JAMA, and DISCERN), and content (25 predefined key facts) of the provided information. Results: The websites were most commonly scientific or governmental (n = 57, 46%), and the median time since the last update was 24 months. Quality was rated with a median JAMA score of 2 (1; 4) and a median overall DISCERN score of 2.4 (1.1; 4.1), both scores represent partially sufficient information. In total, 15% of websites had a HON certificate. Website content measured by the median key fact score was 19 (ranging from 2.5 to 25) with the lowest scores for acute vs. chronic course of the disease, screening for extrapulmonary disease, and diffuse body pain. Poor results were achieved in industry websites and blogs (p = 0.047) with significant differences regarding definition (p = 0.004) and evaluation (p = 0.021). Discussion: Sarcoidosis-related content of internet resources is partially sufficient; however, several important aspects are frequently not addressed, and the quality of information is moderate. Future directions should focus on providing reliable and comprehensive information on sarcoidosis; physicians from different disciplines and patients including self-support groups should collaborate on achieving this.

This article contains a selection of scientific highlights in the field of interstitial lung diseases (ILDs) presented at the hybrid European Respiratory Society International Congress 2022. Early Career Members of Assembly 12 summarise recent advances in translational and clinical research in idiopathic interstitial pneumonias, ILDs of known origin, sarcoidosis and other granulomatous diseases, and rare ILDs. Many studies focused on evaluation of diagnostic and prognostic (bio)markers, and novel pharmacological and nonpharmacological treatment options for different ILDs. In addition, new insights in clinical, physiological and radiological features of various rare ILDs were presented.

Importance: There is a major need for effective, well-tolerated treatments for idiopathic pulmonary fibrosis (IPF). Objective: To assess the efficacy and safety of the autotaxin inhibitor ziritaxestat in patients with IPF. Design, setting, and participants: The 2 identically designed, phase 3, randomized clinical trials, ISABELA 1 and ISABELA 2, were conducted in Africa, Asia-Pacific region, Europe, Latin America, the Middle East, and North America (26 countries). A total of 1306 patients with IPF were randomized (525 patients at 106 sites in ISABELA 1 and 781 patients at 121 sites in ISABELA 2). Enrollment began in November 2018 in both trials and follow-up was completed early due to study termination on April 12, 2021, for ISABELA 1 and on March 30, 2021, for ISABELA 2. Interventions: Patients were randomized 1:1:1 to receive 600 mg of oral ziritaxestat, 200 mg of ziritaxestat, or placebo once daily in addition to local standard of care (pirfenidone, nintedanib, or neither) for at least 52 weeks. Main outcomes and measures: The primary outcome was the annual rate of decline for forced vital capacity (FVC) at week 52. The key secondary outcomes were disease progression, time to first respiratory-related hospitalization, and change from baseline in St George's Respiratory Questionnaire total score (range, 0 to 100; higher scores indicate poorer health-related quality of life). Results: At the time of study termination, 525 patients were randomized in ISABELA 1 and 781 patients in ISABELA 2 (mean age: 70.0 [SD, 7.2] years in ISABELA 1 and 69.8 [SD, 7.1] years in ISABELA 2; male: 82.4% and 81.2%, respectively). The trials were terminated early after an independent data and safety monitoring committee concluded that the benefit to risk profile of ziritaxestat no longer supported their continuation. Ziritaxestat did not improve the annual rate of FVC decline vs placebo in either study. In ISABELA 1, the least-squares mean annual rate of FVC decline was -124.6 mL (95% CI

Advanced large language models like ChatGPT have gained considerable attention recently, including among students. However, while the debate on ChatGPT in academia is making waves, more understanding is needed among lecturers and teachers on how students use and perceive ChatGPT. To address this gap, we analyzed the content on ChatGPT available on TikTok in February 2023. TikTok is a rapidly growing social media platform popular among individuals under 30. Specifically, we analyzed the content of the 100 most popular videos in English tagged with #chatgpt, which collectively garnered over 250 million views. Most of the videos we studied promoted the use of ChatGPT for tasks like writing essays or code. In addition, many videos discussed AI detectors, with a focus on how other tools can help to transform ChatGPT output to fool these detectors. This also mirrors the discussion among educators on how to treat ChatGPT as lecturers and teachers in teaching and grading. What is, however, missing from the analyzed clips on TikTok are videos that discuss ChatGPT producing content that is nonsensical or unfaithful to the training data.

Background: Pulmonary fibrosis (PF) is caused by a heterogeneous group of diseases, with a high inter-individual variability in disease trajectory. Identifying disease progression in patients with PF has impact on clinical management decisions. However, strategies to early identify and predict disease progression for these patients are currently lacking. In this study, we aim to assess long-term FVC change in patients with PF measured with home spirometry, and evaluate the feasibility of a multinational patient-led registry in PF. In addition, we will assess validity of patient-reported outcomes (PROMs) for the different subgroups of patients with PF. Methods: In this international, prospective, multicenter, observational study, we aim to include 700 patients across seven European countries. Patients will monitor their disease course for a period of two years using an online home monitoring program (I-FILE), which includes home spirometry, pulse oximetry, and PROMs. Results will be directly sent to the hospital via the online application. Patients will be asked to perform daily home spirometry and pulse oximetry in the first three months, followed by once weekly measurements for a period of two years. PROMs will be completed in the online I-FILE application every six months, including the King’s brief Interstitial Lung Disease Health Status, The EuroQol five dimensions five-level, Visual Analogue Scales on cough, dyspnea, fatigue and general complaints, Leicester Cough Questionnaire, Fatigue Assessment Scale, Work Productivity and Activity Impairment Questionnaire, Global Rating of Change Scale, and Living with Pulmonary Fibrosis questionnaire. Discussion: This study will provide much needed insights in disease trajectories of the different subgroups of patients with PF. Simultaneously, the I-FILE study will yield valuable information on the use and feasibility of home-based data collection. This international patient-led registry will facilitate trans-border col

The world of rare interstitial lung diseases (ILDs) is diverse and complex. Diagnosis and therapy usually pose challenges. This review describes a selection of rare and ultrarare ILDs including pulmonary alveolar proteinosis, pulmonary alveolar microlithiasis and pleuroparenchymal fibroelastosis. In addition, monogenic ILDs or ILDs in congenital syndromes and various multiple cystic lung diseases will be discussed. All these conditions are part of the scope of the European Reference Network on rare respiratory diseases (ERN-LUNG). Epidemiology, pathogenesis, diagnostics and treatment of each disease are presented.

Machine Learning and Deep Learning have achieved an impressive standard today, enabling us to answer questions that were inconceivable a few years ago. Besides these successes, it becomes clear, that beyond pure prediction, which is the primary strength of most supervised machine learning algorithms, the quantification of uncertainty is relevant and necessary as well. While first concepts and ideas in this direction have emerged in recent years, this paper adopts a conceptual perspective and examines possible sources of uncertainty. By adopting the viewpoint of a statistician, we discuss the concepts of aleatoric and epistemic uncertainty, which are more commonly associated with machine learning. The paper aims to formalize the two types of uncertainty and demonstrates that sources of uncertainty are miscellaneous and can not always be decomposed into aleatoric and epistemic. Drawing parallels between statistical concepts and uncertainty in machine learning, we also demonstrate the role of data and their influence on uncertainty.

Survey research aims to collect robust and reliable data from respondents. However, despite researchers’ efforts in designing questionnaires, survey instruments may be imperfect, and question structure not as clear as could be, thus creating a burden for respondents. If it were possible to detect such problems, this knowledge could be used to predict problems in a questionnaire during pretesting, inform real-time interventions through responsive questionnaire design, or to indicate and correct measurement error after the fact. Previous research has used paradata, specifically response times, to detect difficulties and help improve user experience and data quality. Today, richer data sources are available, for example, movements respondents make with their mouse, as an additional detailed indicator for the respondent–survey interaction. This article uses machine learning techniques to explore the predictive value of mouse-tracking data regarding a question’s difficulty. We use data from a survey on respondents’ employment history and demographic information, in which we experimentally manipulate the difficulty of several questions. Using measures derived from mouse movements, we predict whether respondents have answered the easy or difficult version of a question, using and comparing several state-of-the-art supervised learning methods. We have also developed a personalization method that adjusts for respondents’ baseline mouse behavior and evaluate its performance. For all three manipulated survey questions, we find that including the full set of mouse movement measures and accounting for individual differences in these measures improve prediction performance over response-time-only models., German Research Foundation (DFG), Peer Reviewed

Survey research aims to collect robust and reliable data from respondents. However, despite researchers’ efforts in designing questionnaires, survey instruments may be imperfect, and question structure not as clear as could be, thus creating a burden for respondents. If it were possible to detect such problems, this knowledge could be used to predict problems in a questionnaire during pretesting, inform real-time interventions through responsive questionnaire design, or to indicate and correct measurement error after the fact. Previous research has used paradata, specifically response times, to detect difficulties and help improve user experience and data quality. Today, richer data sources are available, for example, movements respondents make with their mouse, as an additional detailed indicator for the respondent–survey interaction. This article uses machine learning techniques to explore the predictive value of mouse-tracking data regarding a question’s difficulty. We use data from a survey on respondents’ employment history and demographic information, in which we experimentally manipulate the difficulty of several questions. Using measures derived from mouse movements, we predict whether respondents have answered the easy or difficult version of a question, using and comparing several state-of-the-art supervised learning methods. We have also developed a personalization method that adjusts for respondents’ baseline mouse behavior and evaluate its performance. For all three manipulated survey questions, we find that including the full set of mouse movement measures and accounting for individual differences in these measures improve prediction performance over response-time-only models., German Research Foundation (DFG), Peer Reviewed

When training data are collected from human annotators, the design of the annotation instrument, the instructions given to annotators, the characteristics of the annotators, and their interactions can impact training data. This study demonstrates that design choices made when creating an annotation instrument also impact the models trained on the resulting annotations. We introduce the term annotation sensitivity to refer to the impact of annotation data collection methods on the annotations themselves and on downstream model performance and predictions. We collect annotations of hate speech and offensive language in five experimental conditions of an annotation instrument, randomly assigning annotators to conditions. We then fine-tune BERT models on each of the five resulting datasets and evaluate model performance on a holdout portion of each condition. We find considerable differences between the conditions for 1) the share of hate speech/offensive language annotations, 2) model performance, 3) model predictions, and 4) model learning curves. Our results emphasize the crucial role played by the annotation instrument which has received little attention in the machine learning literature. We call for additional research into how and why the instrument impacts the annotations to inform the development of best practices in instrument design., Comment: EMNLP 2023 Findings: https://aclanthology.org/2023.findings-emnlp.992

Machine Learning (ML) systems are becoming instrumental in the public sector, with applications spanning areas like criminal justice, social welfare, financial fraud detection, and public health. While these systems offer great potential benefits to institutional decision-making processes, such as improved efficiency and reliability, they still face the challenge of aligning nuanced policy objectives with the precise formalization requirements necessitated by ML models. In this paper, we aim to bridge the gap between ML model requirements and public sector decision-making by presenting a comprehensive overview of key technical challenges where disjunctions between policy goals and ML models commonly arise. We concentrate on pivotal points of the ML pipeline that connect the model to its operational environment, discussing the significance of representative training data and highlighting the importance of a model setup that facilitates effective decision-making. Additionally, we link these challenges with emerging methodological advancements, encompassing causal ML, domain adaptation, uncertainty quantification, and multi-objective optimization, illustrating the path forward for harmonizing ML and public sector objectives.

For healthcare datasets, it is often not possible to combine data samples from multiple sites due to ethical, privacy or logistical concerns. Federated learning allows for the utilisation of powerful machine learning algorithms without requiring the pooling of data. Healthcare data has many simultaneous challenges which require new methodologies to address, such as highly-siloed data, class imbalance, missing data, distribution shifts and non-standardised variables. Federated learning adds significant methodological complexity to conventional centralised machine learning, requiring distributed optimisation, communication between nodes, aggregation of models and redistribution of models. In this systematic review, we consider all papers on Scopus that were published between January 2015 and February 2023 and which describe new federated learning methodologies for addressing challenges with healthcare data. We performed a detailed review of the 89 papers which fulfilled these criteria. Significant systemic issues were identified throughout the literature which compromise the methodologies in many of the papers reviewed. We give detailed recommendations to help improve the quality of the methodology development for federated learning in healthcare., Comment: Supplementary table of extracted data at the end of the document

Although the NLP community has adopted central differential privacy as a go-to framework for privacy-preserving model training or data sharing, the choice and interpretation of the key parameter, privacy budget $\varepsilon$ that governs the strength of privacy protection, remains largely arbitrary. We argue that determining the $\varepsilon$ value should not be solely in the hands of researchers or system developers, but must also take into account the actual people who share their potentially sensitive data. In other words: Would you share your instant messages for $\varepsilon$ of 10? We address this research gap by designing, implementing, and conducting a behavioral experiment (311 lay participants) to study the behavior of people in uncertain decision-making situations with respect to privacy-threatening situations. Framing the risk perception in terms of two realistic NLP scenarios and using a vignette behavioral study help us determine what $\varepsilon$ thresholds would lead lay people to be willing to share sensitive textual data - to our knowledge, the first study of its kind., Comment: Accepted at LREC-COLING 2024; final camera-ready version

Clinical data is often affected by clinically irrelevant factors such as discrepancies between measurement devices or differing processing methods between sites. In the field of machine learning (ML), these factors are known as domains and the distribution differences they cause in the data are known as domain shifts. ML models trained using data from one domain often perform poorly when applied to data from another domain, potentially leading to wrong predictions. As such, developing machine learning models that can generalise well across multiple domains is a challenging yet essential task in the successful application of ML in clinical practice. In this paper, we propose a novel disentangled autoencoder (Dis-AE) neural network architecture that can learn domain-invariant data representations for multi-label classification of medical measurements even when the data is influenced by multiple interacting domain shifts at once. The model utilises adversarial training to produce data representations from which the domain can no longer be determined. We evaluate the model's domain generalisation capabilities on synthetic datasets and full blood count (FBC) data from blood donors as well as primary and secondary care patients, showing that Dis-AE improves model generalisation on multiple domains simultaneously while preserving clinically relevant information., Comment: 17 pages main body, 5 figures, 18 pages of appendix

Rationale: There is an urgent need for simple, cost-effective prognostic biomarkers for idiopathic pulmonary fibrosis (IPF); biomarkers that show potential include monocyte count. Objectives: We used pooled data from pirfenidone and IFNγ-1b trials to explore the association between monocyte count and prognosis in patients with IPF. Methods: This retrospective pooled analysis included patients (active and placebo arms) from the following four phase III, randomized, placebo-controlled trials: ASCEND (NCT01366209), CAPACITY (NCT00287729 and NCT00287716), and INSPIRE (NCT00075998). Outcomes included IPF progression (≥10% absolute decline in FVC% predicted, ≥50 m decline in 6-minute-walk distance, or death), all-cause hospitalization, and all-cause mortality over 1 year. The relationship between monocyte count (defined as time-dependent) and outcomes was assessed using bivariate and multivariable models. Measurements and Main Results: This analysis included 2,067 patients stratified by monocyte count (at baseline: <0.60 × 109 cells/L [n = 1,609], 0.60 to <0.95 × 109 cells/L [n = 408], and ≥0.95 × 109 cells/L [n = 50]). In adjusted analyses, a higher proportion of patients with monocyte counts of 0.60 to <0.95 × 109 cells/L or ≥0.95 × 109 cells/L versus <0.60 × 109 cells/L experienced IPF progression (P = 0.016 and P = 0.002, respectively), all-cause hospitalization (P = 0.030 and P = 0.003, respectively), and all-cause mortality (P = 0.005 and P < 0.001, respectively) over 1 year. Change in monocyte count from baseline was not associated with any of the outcomes over 1 year and did not appear to be affected by study treatment. Conclusions: In patients with IPF, elevated monocyte count was associated with increased risks of IPF progression, hospitalization, and mortality. Monocyte count may provide a simple and inexpensive prognostic biomarker in IPF.

Idiopathic pulmonary fibrosis (IPF) is a chronic and progressive disorder with an estimated median survival time of 3–5 years after diagnosis. This condition occurs primarily in elderly subjects, and epidemiological studies suggest that the main risk factors, ageing and exposure to cigarette smoke, are associated with both pulmonary and extrapulmonary comorbidities (defined as the occurrence of two or more disorders in a single individual). Ageing and senescence, through interactions with environmental factors, may contribute to the pathogenesis of IPF by various mechanisms, causing lung epithelium damage and increasing the resistance of myofibroblasts to apoptosis, eventually resulting in extracellular matrix accumulation and pulmonary fibrosis. As a paradigm, syndromes featuring short telomeres represent archetypal premature ageing syndromes and are often associated with pulmonary fibrosis. The pathophysiological features induced by ageing and senescence in patients with IPF may translate to pulmonary and extrapulmonary features, including emphysema, pulmonary hypertension, lung cancer, coronary artery disease, gastro-oesophageal reflux, diabetes mellitus and many other chronic diseases, which may lead to substantial negative consequences in terms of various outcome parameters in IPF. Therefore, the careful diagnosis and treatment of comorbidities may represent an outstanding chance to improve quality of life and survival, and it is necessary to contemplate all possible management options for IPF, including early identification and treatment of comorbidities.