Comparing Commercial and Open-Source Large Language Models for Labeling Chest Radiograph Reports.

Background Rapid advances in large language models (LLMs) have led to the development of numerous commercial and open-source models. While recent publications have explored OpenAI's GPT-4 to extract information of interest from radiology reports, there has not been a real-world comparison of GPT-4 to leading open-source models. Purpose To compare different leading open-source LLMs to GPT-4 on the task of extracting relevant findings from chest radiograph reports. Materials and Methods Two independent datasets of free-text radiology reports from chest radiograph examinations were used in this retrospective study performed between February 2, 2024, and February 14, 2024. The first dataset consisted of reports from the ImaGenome dataset, providing reference standard annotations from the MIMIC-CXR database acquired between 2011 and 2016. The second dataset consisted of randomly selected reports created at the Massachusetts General Hospital between July 2019 and July 2021. In both datasets, the commercial models GPT-3.5 Turbo and GPT-4 were compared with open-source models that included Mistral-7B and Mixtral-8 × 7B (Mistral AI), Llama 2-13B and Llama 2-70B (Meta), and Qwen1.5-72B (Alibaba Group), as well as CheXbert and CheXpert-labeler (Stanford ML Group), in their ability to accurately label the presence of multiple findings in radiograph text reports using zero-shot and few-shot prompting. The McNemar test was used to compare F1 scores between models. Results On the ImaGenome dataset (n = 450), the open-source model with the highest score, Llama 2-70B, achieved micro F1 scores of 0.97 and 0.97 for zero-shot and few-shot prompting, respectively, compared with the GPT-4 F1 scores of 0.98 and 0.98 (P > .99 and < .001 for superiority of GPT-4). On the institutional dataset (n = 500), the open-source model with the highest score, an ensemble model, achieved micro F1 scores of 0.96 and 0.97 for zero-shot and few-shot prompting, respectively, compared with the GPT-4 F1 scores of 0.98 and 0.97 (P < .001 and > .99 for superiority of GPT-4). Conclusion Although GPT-4 was superior to open-source models in zero-shot report labeling, few-shot prompting with a small number of example reports closely matched the performance of GPT-4. The benefit of few-shot prompting varied across datasets and models. © RSNA, 2024 Supplemental material is available for this article.

Correlated rotational alignment spectroscopy: a new tool for high-resolution spectroscopy and the analysis of heterogeneous samples.

Correlated rotational alignment spectroscopy correlates observables of ultrafast gas-phase spectroscopy with high-resolution, broad-band rotational Raman spectra. This article reviews the measurement principle of CRASY, existing implementations for mass-correlated measurements, and the potential for future developments. New spectroscopic capabilities are discussed in detail: signals for individual sample components can be separated even in highly heterogeneous samples. Isotopologue rotational spectra can be observed at natural isotope abundance. Fragmentation channels are readily assigned in molecular and cluster mass spectra. And finally, rotational Raman spectra can be measured with sub-MHz resolution, an improvement of several orders-of-magnitude as compared to preceding experiments.

Myddosomes in Toll-like receptor signaling-one to bind and rule them all.

Toll-like receptors (TLRs) are innate immune sensors for the presence of pathogens and endogenous danger signals. TLR activation results in conserved intracellular signaling events that orchestrate inflammation and antimicrobial defense. While the identity and interplay of key TLR signaling components are well established, how these largely cytosolic proteins are physically connected is not well understood. For the activation of conserved intracellular signaling events, most TLRs engage the adapter MyD88 (myeloid differentiation primary response 88), which assembles into higher-order protein complexes, myddosomes. In their recent publication, Fisch et al. present evidence that oligomeric myddosomes detach from initiating TLRs and evolve into larger scaffolds that dynamically assemble not only proximal but also distal cytosolic elements required to execute the entire cascade of the TLR-MyD88 signaling pathway. Coinciding with decline in TLR signaling over time, myddosomes progressively recruit autophagy machinery that mediates myddosome clearance. These findings expand the current understanding of TLR signaling by positioning myddosomes as the central structural element that physically assembles the key executors and regulators of TLR-MyD88-dependent intracellular signaling cascades.

Public Perspectives on Direct-to-Consumer Testing Oversight.

The FDA published a final rule for Medical Devices; Laboratory Developed Tests in the Federal Register on May 6, 2024, which aims to ensure the safety and effectiveness of laboratory developed tests (LDTs) by amending current regulations. The rule also includes a policy to phase out the FDA's general enforcement discretion approach for LDTs, aligning them with other In Vitro Diagnostic Devices. Notably, direct-to-consumer (DTC) testing is exempt from this policy shift, as the FDA believes this category of tests has already met applicable requirements. This rule was first proposed in the Federal Register on October 3, 2023. The publication of this proposed rule sparked a considerable volume of public reactions during the comment period of the rule-making process, comprising general sentiment, key concerns, and suggestions. This commentary analyzes these concerns, particularly focusing on DTC tests, and offers recommendations, including reassessing the FDA's enforcement discretion for hybrid DTC tests, advocating for clear guidance on clinical oversight, and prioritizing a risk-based enforcement approach. Additionally, enhancing public education about the risks of DTC testing is crucial for safeguarding public health.

Validation of the German version of the Asthma Impairment and Risk Questionnaire (AIRQ).

BACKGROUND: The Asthma Impairment and Risk Questionnaire (AIRQ), a 10-item, equally weighted, yes/no tool assessing symptom impairment and risk of exacerbations in patients with asthma aged ≥12 years, was developed and validated in a US patient population to evaluate varying levels of asthma control. This study aimed to validate the German language version of the AIRQ in patients aged ≥12 years with different levels of asthma control. METHODS: A cross-sectional, observational, multi-centre study comprising a single visit was conducted in multiple specialised asthma centres and general practices in Germany. A total of 300 patients completed the following measures: 1) Patient Sociodemographic and Clinical Questionnaire, 2) AIRQ, 3) Asthma Control Test (ACT), and 4) Asthma Control Questionnaire (ACQ-6). Logistic regression analyses were conducted to assess the AIRQ score cut points with the greatest predictive validity in discriminating between different control levels relative to a standard of ACT plus prior-year exacerbations or ACQ-6 plus prior-year exacerbations. RESULTS: The German version of the AIRQ demonstrated a robust capability to correctly identify well-controlled versus not well- or very poorly controlled (AUC values of 0.90 or higher) and well- or not well-controlled versus very poorly controlled asthma (AUC values of 0.89 or higher). CONCLUSIONS: The German version of the AIRQ is a suitable tool to identify adults with varying levels of asthma control, which in turn can help to accurately identify patients with uncontrolled asthma in clinical practice.

Assessing Laterality Errors in Radiology: Comparing Generative Artificial Intelligence and Natural Language Processing.

PURPOSE: We compared the performance of generative artificial intelligence (AI) (Augmented Transformer Assisted Radiology Intelligence [ATARI, Microsoft Nuance, Microsoft Corporation, Redmond, Washington]) and natural language processing (NLP) tools for identifying laterality errors in radiology reports and images. METHODS: We used an NLP-based (mPower, Microsoft Nuance) tool to identify radiology reports flagged for laterality errors in its Quality Assurance Dashboard. The NLP model detects and highlights laterality mismatches in radiology reports. From an initial pool of 1,124 radiology reports flagged by the NLP for laterality errors, we selected and evaluated 898 reports that encompassed radiography, CT, MRI, and ultrasound modalities to ensure comprehensive coverage. A radiologist reviewed each radiology report to assess if the flagged laterality errors were present (reporting error-true-positive) or absent (NLP error-false-positive). Next, we applied ATARI to 237 radiology reports and images with consecutive NLP true-positive (118 reports) and false-positive (119 reports) laterality errors. We estimated accuracy of NLP and generative AI tools to identify overall and modality-wise laterality errors. RESULTS: Among the 898 NLP-flagged laterality errors, 64% (574 of 898) had NLP errors and 36% (324 of 898) were reporting errors. The text query ATARI feature correctly identified the absence of laterality mismatch (NLP false-positives) with a 97.4% accuracy (115 of 118 reports; 95% confidence interval [CI] = 96.5%-98.3%). Combined vision and text query resulted in 98.3% accuracy (116 of 118 reports or images; 95% CI = 97.6%-99.0%), and query alone had a 98.3% accuracy (116 of 118 images; 95% CI = 97.6%-99.0%). CONCLUSION: The generative AI-empowered ATARI prototype outperformed the assessed NLP tool for determining true and false laterality errors in radiology reports while enabling an image-based laterality determination. Underlying errors in ATARI text query in complex radiology reports emphasize the need for further improvement in the technology.

Heme d formation in a Shewanella benthica hemoglobin.

In our continued investigations of microbial globins, we solved the structure of a truncated hemoglobin from Shewanella benthica, an obligate psychropiezophilic bacterium. The distal side of the heme active site is lined mostly with hydrophobic residues, with the exception of a tyrosine, Tyr34 (CD1) and a histidine, His24 (B13). We found that purified SbHbN, when crystallized in the ferric form with polyethylene glycol as precipitant, turned into a green color over weeks. The electron density obtained from the green crystals accommodated a trans heme d, a chlorin-type derivative featuring a γ-spirolactone and a vicinal hydroxyl group on a pyrroline ring. In solution, exposure of the protein to one equivalent of hydrogen peroxide resulted in a similar green color change, but caused by the formation of multiple products. These were oxidation species released on protein denaturation, likely including heme d, and a species with heme covalently attached to the polypeptide. The Tyr34Phe replacement prevented the formation of both heme d and the covalent linkage. The ready modification of heme b by SbHbN expands the range of chemistries supported by the globin fold and offers a route to a novel heme cofactor.

No code machine learning: validating the approach on use-case for classifying clavicle fractures.

PURPOSE: We created an infrastructure for no code machine learning (NML) platform for non-programming physicians to create NML model. We tested the platform by creating an NML model for classifying radiographs for the presence and absence of clavicle fractures. METHODS: Our IRB-approved retrospective study included 4135 clavicle radiographs from 2039 patients (mean age 52 ± 20 years, F:M 1022:1017) from 13 hospitals. Each patient had two-view clavicle radiographs with axial and anterior-posterior projections. The positive radiographs had either displaced or non-displaced clavicle fractures. We configured the NML platform to automatically retrieve the eligible exams using the series' unique identification from the hospital virtual network archive via web access to DICOM Objects. The platform trained a model until the validation loss plateaus. Once the testing was complete, the platform provided the receiver operating characteristics curve and confusion matrix for estimating sensitivity, specificity, and accuracy. RESULTS: The NML platform successfully retrieved 3917 radiographs (3917/4135, 94.7 %) and parsed them for creating a ML classifier with 2151 radiographs in the training, 100 radiographs for validation, and 1666 radiographs in testing datasets (772 radiographs with clavicle fracture, 894 without clavicle fracture). The network identified clavicle fracture with 90 % sensitivity, 87 % specificity, and 88 % accuracy with AUC of 0.95 (confidence interval 0.94-0.96). CONCLUSION: A NML platform can help physicians create and test machine learning models from multicenter imaging datasets such as the one in our study for classifying radiographs based on the presence of clavicle fracture.

Artificial Intelligence in Cataract Surgery: A Systematic Review.

Purpose: The purpose of this study was to assess the current use and reliability of artificial intelligence (AI)-based algorithms for analyzing cataract surgery videos. Methods: A systematic review of the literature about intra-operative analysis of cataract surgery videos with machine learning techniques was performed. Cataract diagnosis and detection algorithms were excluded. Resulting algorithms were compared, descriptively analyzed, and metrics summarized or visually reported. The reproducibility and reliability of the methods and results were assessed using a modified version of the Medical Image Computing and Computer-Assisted (MICCAI) checklist. Results: Thirty-eight of the 550 screened studies were included, 20 addressed the challenge of instrument detection or tracking, 9 focused on phase discrimination, and 8 predicted skill and complications. Instrument detection achieves an area under the receiver operator characteristic curve (ROC AUC) between 0.976 and 0.998, instrument tracking an mAP between 0.685 and 0.929, phase recognition an ROC AUC between 0.773 and 0.990, and complications or surgical skill performs with an ROC AUC between 0.570 and 0.970. Conclusions: The studies showed a wide variation in quality and pose a challenge regarding replication due to a small number of public datasets (none for manual small incision cataract surgery) and seldom published source code. There is no standard for reported outcome metrics and validation of the models on external datasets is rare making comparisons difficult. The data suggests that tracking of instruments and phase detection work well but surgical skill and complication recognition remains a challenge for deep learning. Translational Relevance: This overview of cataract surgery analysis with AI models provides translational value for improving training of the clinician by identifying successes and challenges.

Patient-Centric Quality Standards.

To ensure the quality, safety and efficacy of medicinal products, it is necessary to develop and execute appropriate manufacturing process and product control strategies. Traditionally, product control strategies have focused on testing known quality attributes with limits derived from levels administered in preclinical and clinical studies with an associated statistical analysis to account for variability. However, not all quality attributes have impact to the patient and those with the potential to impact safety and efficacy may not be significant when dosed at patient-centric levels. Therefore, achieving patient-centricity is understanding patient relevance, which is defined as the level of impact that a quality attribute could have on safety and efficacy within the potential exposure range. A patient-centric quality standard (PCQS) is therefore a set of patient relevant attributes and their associated acceptance ranges to which a drug product should conform within the expected patient exposure range. This manuscript describes historical perspectives details the way to create and leverage a PCQS in a variety of pharmaceutical product modalities.

Veteran influenza vaccination acceptance rates after completion of the COVID-19 vaccination series among historical influenza vaccine refusers.

BACKGROUND: Vaccine hesitancy remains an obstacle in disease prevention. The recent COVID-19 pandemic highlighted this issue and may influence acceptance of other recommended immunizations. The objective of this study was to determine the association between receiving the COVID-19 vaccination and the subsequent acceptance of the influenza vaccination in a Veteran population that historically declined influenza vaccination. METHODS: Influenza vaccination acceptance rates for the 2021-2022 influenza season were compared in patients who historically declined the influenza vaccine and either received or declined COVID-19 vaccinations. Logistic regression analysis was used to analyze factors associated with receiving influenza vaccination among vaccine hesitant individuals. RESULTS: A higher proportion of patients who had received the COVID-19 vaccination(s) subsequently accepted the influenza vaccination compared to the control group (37% vs. 11%, OR = 5.03; CI 3.15-8.26; p = 0.0001). CONCLUSION: Among previous influenza vaccine decliners, those who received COVID-19 vaccination had significantly higher odds of receiving subsequent influenza vaccination.

Accurate drusen segmentation in optical coherence tomography via order-constrained regression of retinal layer heights.

Drusen are an important biomarker for age-related macular degeneration (AMD). Their accurate segmentation based on optical coherence tomography (OCT) is therefore relevant to the detection, staging, and treatment of disease. Since manual OCT segmentation is resource-consuming and has low reproducibility, automatic techniques are required. In this work, we introduce a novel deep learning based architecture that directly predicts the position of layers in OCT and guarantees their correct order, achieving state-of-the-art results for retinal layer segmentation. In particular, the average absolute distance between our model's prediction and the ground truth layer segmentation in an AMD dataset is 0.63, 0.85, and 0.44 pixel for Bruch's membrane (BM), retinal pigment epithelium (RPE) and ellipsoid zone (EZ), respectively. Based on layer positions, we further quantify drusen load with excellent accuracy, achieving 0.994 and 0.988 Pearson correlation between drusen volumes estimated by our method and two human readers, and increasing the Dice score to 0.71 ± 0.16 (from 0.60 ± 0.23) and 0.62 ± 0.23 (from 0.53 ± 0.25), respectively, compared to a previous state-of-the-art method. Given its reproducible, accurate, and scalable results, our method can be used for the large-scale analysis of OCT data.

Spatially regularized low-rank tensor approximation for accurate and fast tractography.

Tractography based on diffusion Magnetic Resonance Imaging (dMRI) is the prevalent approach to the in vivo delineation of white matter tracts in the human brain. Many tractography methods rely on models of multiple fiber compartments, but the local dMRI information is not always sufficient to reliably estimate the directions of secondary fibers. Therefore, we introduce two novel approaches that use spatial regularization to make multi-fiber tractography more stable. Both represent the fiber Orientation Distribution Function (fODF) as a symmetric fourth-order tensor, and recover multiple fiber orientations via low-rank approximation. Our first approach computes a joint approximation over suitably weighted local neighborhoods with an efficient alternating optimization. The second approach integrates the low-rank approximation into a current state-of-the-art tractography algorithm based on the unscented Kalman filter (UKF). These methods were applied in three different scenarios. First, we demonstrate that they improve tractography even in high-quality data from the Human Connectome Project, and that they maintain useful results with a small fraction of the measurements. Second, on the 2015 ISMRM tractography challenge, they increase overlap, while reducing overreach, compared to low-rank approximation without joint optimization or the traditional UKF, respectively. Finally, our methods permit a more comprehensive reconstruction of tracts surrounding a tumor in a clinical dataset. Overall, both approaches improve reconstruction quality. At the same time, our modified UKF significantly reduces the computational effort compared to its traditional counterpart, and to our joint approximation. However, when used with ROI-based seeding, joint approximation more fully recovers fiber spread.

Addressing the Challenges of Implementing Artificial Intelligence Tools in Clinical Practice: Principles From Experience.

The multitude of artificial intelligence (AI)-based solutions, vendors, and platforms poses a challenging proposition to an already complex clinical radiology practice. Apart from assessing and ensuring acceptable local performance and workflow fit to improve imaging services, AI tools require multiple stakeholders, including clinical, technical, and financial, who collaborate to move potential deployable applications to full clinical deployment in a structured and efficient manner. Postdeployment monitoring and surveillance of such tools require an infrastructure that ensures proper and safe use. Herein, the authors describe their experience and framework for implementing and supporting the use of AI applications in radiology workflow.

Distinct contributions of the innate immune receptors TLR2 and RP105 to formation and architecture of structured lung granulomas in mice infected with Mycobacterium tuberculosis.

Granulomas are key histopathological features of Mycobacterium tuberculosis (Mtb) infection, with complex roles in pathogen control and dissemination. Thus, understanding drivers and regulators of granuloma formation is important for improving tuberculosis diagnosis, treatment, and prevention. Yet, molecular mechanisms underpinning granuloma formation and dynamics remain poorly understood. Here we used low-dose Mtb infection of C57BL/6 mice, which elicits structured lung granulomas composed of central macrophage clusters encased by a lymphocyte mantle, alongside the disorganized lymphocyte and macrophage clusters commonly observed in Mtb-infected mice. Using gene-deficient mice, we observed that Toll-like receptor (TLR) 2 and the TLR-related Radioprotective 105 kDa protein (RP105) contributed to the extent and spatial positioning of pathology in infected lung tissues, consistent with functional cooperation between TLR2 and RP105 in the innate immune recognition of Mtb. In mice infected with the highly virulent Mtb clinical isolate HN878, TLR2, but not RP105, positively regulated the extent of central macrophage regions within structured granulomas. Moreover, RP105, but not TLR2, promoted the formation of structured lung granulomas, suggesting that the functions of RP105 as an innate immune sensor for Mtb reach beyond its roles as TLR2 co-receptor. TLR2 and RP105 contributions to lung pathology are governed by Mtb biology, as neither receptor affected the frequency or architecture of structured granulomas in mice infected with the reference strain Mtb H37Rv. Thus, by revealing distinctive as well as cooperative functions of TLR2 and RP105 in lung pathology, our data identify TLRs as molecular determinants of TB granuloma formation and architecture, and expand understanding of how interactions between innate immune receptors and Mtb shape TB disease manifestation.

Structure of benzene from mass-correlated rotational Raman spectroscopy.

We present high resolution rotational Raman spectra and derived geometry parameters for benzene. Rotational Raman spectra with sub-5 MHz resolution were obtained via high-resolution mass-correlated rotational alignment spectroscopy. Isotopologue spectra for C6H6, 13C-C5H6, C6D6, and 13C-C5D6 were distinguished through their correlated mass information. Spectra for 13C6H6 were obtained with lower resolution. Equilibrium and effective bond lengths were estimated from measured inertial moments, based on explicit assumptions and approximations. We discuss the origin of significant bias in previously published geometry parameters and the possibility to derive H,D isotope-specific bond lengths from purely experimental data.

Reducing instability of inter-subject covariance of FDG uptake networks using structure-weighted sparse estimation approach.

PURPOSE: Sparse inverse covariance estimation (SICE) is increasingly utilized to estimate inter-subject covariance of FDG uptake (FDGcov) as proxy of metabolic brain connectivity. However, this statistical method suffers from the lack of robustness in the connectivity estimation. Patterns of FDGcov were observed to be spatially similar with patterns of structural connectivity as obtained from DTI imaging. Based on this similarity, we propose to regularize the sparse estimation of FDGcov using the structural connectivity. METHODS: We retrospectively analyzed the FDG-PET and DTI data of 26 healthy controls, 41 patients with Alzheimer's disease (AD), and 30 patients with frontotemporal lobar degeneration (FTLD). Structural connectivity matrix derived from DTI data was introduced as a regularization parameter to assign individual penalties to each potential metabolic connectivity. Leave-one-out cross validation experiments were performed to assess the differential diagnosis ability of structure weighted SICE approach. A few approaches of structure weighted were compared with the standard SICE. RESULTS: Compared to the standard SICE, structural weighting has shown more stable performance in the supervised classification, especially in the differentiation AD vs. FTLD (accuracy of 89-90%, while unweighted SICE only 85%). There was a significant positive relationship between the minimum number of metabolic connection and the robustness of the classification accuracy (r = 0.57, P < 0.001). Shuffling experiments showed significant differences between classification score derived with true structural weighting and those obtained by randomized structure (P < 0.05). CONCLUSION: The structure-weighted sparse estimation can enhance the robustness of metabolic connectivity, which may consequently improve the differentiation of pathological phenotypes.

Mass-Correlated High-Resolution Spectra and the Structure of Benzene.

Mass-correlated rotational alignment spectroscopy resolved the rotational Raman spectra for 5 benzene isotopologues with unprecedented resolution. 13-C isotopologues were characterized at natural abundance. Fitted rotational constants allowed the analysis of effective and equilibrium bond lengths for benzene with sub-mÅ uncertainties. We found that previously reported experimental structures were wrong by multiple mÅ, due to unrecognized H/D isotope effects. Our results also refute recent experimental and theoretical literature claims of identical effective C-H and C-D bond lengths in benzene and reveal an isotope effect similar to that in other small molecules.

Engaging the Wisdom of Older Veterans to Enhance VA Healthcare, Research, and Services.

BACKGROUND: Stakeholder engagement helps ensure that research is relevant, clinical innovations are responsive, and healthcare services are patient-centered. OBJECTIVE: Establish and sustain a Veteran engagement board involving older Veterans and caregivers to provide input on aging-related research and clinical demonstration projects. DESIGN AND PARTICIPANTS: The Older Veteran Engagement Team (OVET)-a group of eight Veterans and one caregiver who range in age from 62 to 92-was formed in November 2017 and has met monthly since January 2018. The OVET provides feedback on topics that reflect the foci of the VA Eastern Colorado Geriatric Research Education and Clinical Center (GRECC) (e.g., physical functioning, hearing health, and emotional wellness/mental health). Ongoing evaluation documents the return on investment of Veteran engagement. MAIN MEASURES: The OVET member and provider/investigator meeting evaluations with longitudinal follow-up at 6 and 12 months. RESULTS: Return on investment of Veteran engagement is multi-faceted. For OVET, ROI ranges from grant support to improved healthcare quality/efficiency to social-emotional benefits. To date, funding awards total over $2.3 M for NIH and VA-funded projects to which OVET provided substantive feedback. Documented impacts on healthcare services include reductions in patient wait times, more appropriate utilization of services and increased patient satisfaction. Social-emotional benefits include generativity, as OVET members contribute to improving clinical and community-based supports for other Veterans. The OVET provides an opportunity for older Veterans to share their lived experience with trainees and early career investigators who are preparing for careers serving Veterans. CONCLUSION: The OVET is similar to other established stakeholder engagement groups; team members offer their individual viewpoints at any stage of research, clinical demonstration, or quality improvement projects. The OVET provides a mechanism for the voice of older Veterans and caregivers to shape aspects of individual projects. Importantly, these projects support patient-centered care and promote the characteristics of an age-friendly healthcare system.

[Experiences with digital care of patients with chronic and acute lung diseases during the SARS-CoV-2 pandemic]. / Erfahrungen mit der digitalen Versorgung von Patienten mit chronischen und akuten Lungenerkrankungen während der SARS-CoV-2-Pandemie.

BACKGROUND: Management of patients with respiratory disorders, such as asthma or chronic obstructive pulmonary disease (COPD), became challenging during the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic due to infection prevention measures. To maintain care, a remote monitoring program was initiated, comprising a smartphone app and a Bluetooth spirometry device. OBJECTIVE: To assess patient- and physician-related experience with remote monitoring. MATERIAL AND METHODS: Structured questionnaires were developed to rate experiences from the patient or physician perspective on six-level Likert scales. Interactions between patients and physicians via the digital platform and overall utilization was analyzed. RESULTS: A total of 745 patients with asthma, COPD, post-coronavirus disease 2019 (COVID-19) and other respiratory diseases were enrolled from 31 centers in Germany. Mean follow-up was 49.4 ± 12.6 weeks. Each participant submitted on average 289 measurements. Patient-reported experience with the remote monitoring program was positive, with the highest satisfaction reported for "Experience with home measurement" (1.4 ± 0.5; 99% positive), followed by "Communication/interaction" (1.8 ± 0.9; 83% positive) and "Overall satisfaction with program" (1.8 ± 0.8; 87% positive). In all, 70% reported subjective quality of life improvements related to participation in the program. Physician satisfaction with the program was also high with a mean rating of 2.2 ± 1.2. DISCUSSION: App-based remote monitoring was successfully implemented in routine care during the SARS-CoV­2 pandemic and demonstrated potential for improvements in care. Patient-relevant experience was positive in all dimensions and remote monitoring was well accepted. Physicians who participated in the program also expressed positive experiences, as demonstrated by a high level of interaction with the platform and positive evaluations of effects from the program.