Zebrafish Larvae Microinjection and Automated Fluorescence Microscopy for Studying Klebsiella pneumoniae Infection and the Host Immune Response.

Studying host-pathogen interactions is essential for understanding infectious diseases and developing possible treatments, especially for priority pathogens with increased virulence and antibiotic resistance, such as Klebsiella pneumoniae. Over time, this subject has been approached from different perspectives, often using mammal host models and invasive endpoint measurements (e.g., sacrifice and organ extraction). However, taking advantage of technological advances, it is now possible to follow the infective process by noninvasive visualization in real time, using optically amenable surrogate hosts. In this line, this chapter describes a live-cell imaging approach to monitor the interaction of K. pneumoniae and potentially other bacterial pathogens with zebrafish larvae in vivo. This methodology is based on the microinjection of fluorescent bacteria into the otic vesicle, followed by time-lapse observation by automated fluorescence microscopy with environmental control, monitoring the dynamics of immune cell recruitment, bacterial load, and larvae survival.

Machine Learning for RNA Design: LEARNA.

Machine learning algorithms, and in particular deep learning approaches, have recently garnered attention in the field of molecular biology due to remarkable results. In this chapter, we describe machine learning approaches specifically developed for the design of RNAs, with a focus on the learna_tools Python package, a collection of automated deep reinforcement learning algorithms for secondary structure-based RNA design. We explain the basic concepts of reinforcement learning and its extension, automated reinforcement learning, and outline how these concepts can be successfully applied to the design of RNAs. The chapter is structured to guide through the usage of the different programs with explicit examples, highlighting particular applications of the individual tools.

Detection of egg white allergy in children by specific IgE microarray chemiluminescence immunoassay.

BACKGROUND: Allergen testing has emerged as a pivotal component in prevention and treatment strategies for allergic diseases among children and the utilization of specific IgE (sIgE) through a fully automated chemiluminescent microarray immunoassay (CLMIA) has emerged as a promising trend in the simultaneous detection of multiple allergenic components of children. METHODS: The accuracy and reliability of CLMIA were verified using children's serum samples that concentrated on allergens. the allergens. The clinical diagnostic practicability of CLMIA was assessed through comprehensive evaluations including measurements of the limit of detection (LOD), intra-batch, and inter-batch precision, linearity analysis, the cross-contamination rate, and the concordance rate with the Phadia system. RESULTS: After the optimization process of CLMIA, the LODs for allergens were calculated to be below 0.01 kU/L, demonstrating the high sensitivity of CLMIA. All components exhibited good linearity within the range of 0.1-100.0 kU/L and the coefficient of determinations (R2 > 0.99). The data of intra-batch precision (<10 %) and inter-batch data (<15 %) illustrated the high reproducibility of CLMIA. The cross-contamination rates for allergens (<0.5 %) showed the high accuracy of CLMIA without interfering. The positive concordance rate between CLMIA and the Phadia system exceeds 90 % with a good negative concordance rate (>85 %) and the Kappa coefficients (>0.8), suggesting the close alignment of CLMIA and the Phadia system and showing the satisfactory clinical potential of CLMIA in children's allergy disease. CONCLUSIONS: The application of CLMIA has been promising in allergen testing, especially for detecting multiple allergenic components in children.

A comprehensive approach to evaluate human-machine conflicts in shared steering systems.

The shared control authority between drivers and the steering system may lead to human-machine conflicts, threatening both traffic safety and driving experience of collaborative driving systems. Previous evaluation methods relied on subjective judgment and had a singular set of evaluation criteria, making it challenging to obtain a comprehensive and objective assessment. Therefore, we propose a two-phase novel method that integrates eye-tracking data, electromyography signals and vehicle dynamic features to evaluate human-machine conflicts. Firstly, through driving simulation experiments, the correlations between subjective driving experience and objective indices are analyzed. Strongly correlated indices are screened as the effective criteria. In the second phase, the indices are integrated through sparse principal component analysis (SPCA) to formulate a comprehensive objective measure. Subjective driving experience collected from post-drive questionnaires was applied to examine its effectiveness. The results show that the error between the two sets of data is less than 7%, proving the effectives of the proposed method. This study provides a low-cost, high-efficiency method for evaluating human-machine conflicts, which contributes to the development of safer and more harmonious human-machine collaborative driving.

Automated Drugs Dispensing Systems in Hospitals: a Health Technology Assessment (HTA) Study Across Six European Countries.

Background: Automated Drug Dispensing (ADD) systems are considered to be strategic hospital assets used to reduce errors and enhance economic and organizational sustainability. With regards to efficacy and safety, the literature evidence demonstrates the incremental benefits of centralised or decentralised systems compared to manual dispensing. Analyses about organisational and economic sustainability are still lacking and the present study aims to perform a Health Technology Assessment (HTA), producing multidimensional evidence on the use of ADD systems within hospitals. Methods: In 2023, a comprehensive HTA draws insights from healthcare professionals across six European nations: Italy, France, Germany, the Netherlands, the United Kingdom, and Belgium. This appraisal juxtaposed four drug dispensing scenarios: manual methods, centralized ADD systems, decentralized ADD systems, and integrated solutions employing cutting-edge technologies in both central pharmacies and wards. The study deployed an Activity-Based Costing approach that was combined with a cost-effectiveness and Budget Impact Analysis to evaluate economic impacts. Qualitative questionnaires were implemented to assess ethical, legal, organizational, safety, and efficacy aspects. Results: From a multidimensional perspective, healthcare professionals acknowledged ADD manifold advantages of ADD systems. From an organizational perspective and within a 12-month timeframe, transitioning to automation may face initial challenges that are attributed to potential resistance from professionals and significant investments. However, 36 months past its adoption, automation's superiority over manual methods was recognized. Economically, savings burgeoned from +17.9% in UK to +26.6% in Belgian hospitals that adopted integrated systems in comparison to traditional manual approaches. Conclusion: Compared to traditional methods, implementing ADD systems could improve the logistic management of drug in the hospital setting, thereby enhancing safety and efficacy, streamlining the healthcare professionals' workflow, and bolstering financial stability.

Generation of Backward-Looking Complex Reflections for a Motivational Interviewing-Based Smoking Cessation Chatbot Using GPT-4: Algorithm Development and Validation.

Background: Motivational interviewing (MI) is a therapeutic technique that has been successful in helping smokers reduce smoking but has limited accessibility due to the high cost and low availability of clinicians. To address this, the MIBot project has sought to develop a chatbot that emulates an MI session with a client with the specific goal of moving an ambivalent smoker toward the direction of quitting. One key element of an MI conversation is reflective listening, where a therapist expresses their understanding of what the client has said by uttering a reflection that encourages the client to continue their thought process. Complex reflections link the client's responses to relevant ideas and facts to enhance this contemplation. Backward-looking complex reflections (BLCRs) link the client's most recent response to a relevant selection of the client's previous statements. Our current chatbot can generate complex reflections-but not BLCRs-using large language models (LLMs) such as GPT-2, which allows the generation of unique, human-like messages customized to client responses. Recent advancements in these models, such as the introduction of GPT-4, provide a novel way to generate complex text by feeding the models instructions and conversational history directly, making this a promising approach to generate BLCRs. Objective: This study aims to develop a method to generate BLCRs for an MI-based smoking cessation chatbot and to measure the method's effectiveness. Methods: LLMs such as GPT-4 can be stimulated to produce specific types of responses to their inputs by "asking" them with an English-based description of the desired output. These descriptions are called prompts, and the goal of writing a description that causes an LLM to generate the required output is termed prompt engineering. We evolved an instruction to prompt GPT-4 to generate a BLCR, given the portions of the transcript of the conversation up to the point where the reflection was needed. The approach was tested on 50 previously collected MIBot transcripts of conversations with smokers and was used to generate a total of 150 reflections. The quality of the reflections was rated on a 4-point scale by 3 independent raters to determine whether they met specific criteria for acceptability. Results: Of the 150 generated reflections, 132 (88%) met the level of acceptability. The remaining 18 (12%) had one or more flaws that made them inappropriate as BLCRs. The 3 raters had pairwise agreement on 80% to 88% of these scores. Conclusions: The method presented to generate BLCRs is good enough to be used as one source of reflections in an MI-style conversation but would need an automatic checker to eliminate the unacceptable ones. This work illustrates the power of the new LLMs to generate therapeutic client-specific responses under the command of a language-based specification.

(Semi)automated approaches to data extraction for systematic reviews and meta-analyses in social sciences: A living review.

Background: An abundance of rapidly accumulating scientific evidence presents novel opportunities for researchers and practitioners alike, yet such advantages are often overshadowed by resource demands associated with finding and aggregating a continually expanding body of scientific information. Data extraction activities associated with evidence synthesis have been described as time-consuming to the point of critically limiting the usefulness of research. Across social science disciplines, the use of automation technologies for timely and accurate knowledge synthesis can enhance research translation value, better inform key policy development, and expand the current understanding of human interactions, organizations, and systems. Ongoing developments surrounding automation are highly concentrated in research for evidence-based medicine with limited evidence surrounding tools and techniques applied outside of the clinical research community. The goal of the present study is to extend the automation knowledge base by synthesizing current trends in the application of extraction technologies of key data elements of interest for social scientists. Methods: We report the baseline results of a living systematic review of automated data extraction techniques supporting systematic reviews and meta-analyses in the social sciences. This review follows PRISMA standards for reporting systematic reviews. Results: The baseline review of social science research yielded 23 relevant studies. Conclusions: When considering the process of automating systematic review and meta-analysis information extraction, social science research falls short as compared to clinical research that focuses on automatic processing of information related to the PICO framework. With a few exceptions, most tools were either in the infancy stage and not accessible to applied researchers, were domain specific, or required substantial manual coding of articles before automation could occur. Additionally, few solutions considered extraction of data from tables which is where key data elements reside that social and behavioral scientists analyze.

Semi-automated reproducible target transfer for cardiac radioablation - A multi-center cross-validation study within the RAVENTA trial.

BACKGROUND: Stereotactic arrhythmia radioablation (STAR) is a therapeutic option for ventricular tachycardia (VT) where catheter-based ablation is not feasible or has previously failed. Target definition and its transfer from electro-anatomic maps (EAM) to radiotherapy treatment planning systems (TPS) is challenging and operator-dependent. Software solutions have been developed to register EAM with cardiac CT and semi-automatically transfer 2D target surface data into 3D CT volume coordinates. Results of a cross-validation study of two conceptually different software solutions using data from the RAVENTA trial (NCT03867747) are reported. METHODS: Clinical Target Volumes (CTVs) were created from target regions delineated on EAM using two conceptually different approaches by separate investigators on data of 10 patients, blinded to each other's results. Targets were transferred using 3D-3D registration and 2D-3D registration, respectively. The resulting CTVs were compared in a core-lab using two complementary analysis software packages for structure similarity and geometric characteristics. RESULTS: Volumes and surface areas of the CTVs created by both methods were comparable: 14.88 ± 11.72 ml versus 15.15 ± 11.35 ml and 44.29 ± 33.63 cm2 versus 46.43 ± 35.13 cm2. The Dice-coefficient was 0.84 ± 0.04; median surface-distance and Hausdorff-distance were 0.53 ± 0.37 mm and 6.91 ± 2.26 mm, respectively. The 3D-center-of-mass difference was 3.62 ± 0.99 mm. Geometrical volume similarity was 0.94 ± 0.05 %. CONCLUSION: The STAR targets transferred from EAM to TPS using both software solutions resulted in nearly identical 3D structures. Both solutions can be used for QA (quality assurance) and EAM-to-TPS transfer of STAR-targets. Semi-automated methods could potentially help to avoid mistargeting in STAR and offer standardized workflows for methodically harmonized treatments.

Clinical implementation of deep learning robust IMPT planning in oropharyngeal cancer patients: A blinded clinical study.

BACKGROUND AND PURPOSE: This study aimed to evaluate the plan quality of our deep learning-based automated treatment planning method for robustly optimized intensity-modulated proton therapy (IMPT) plans in patients with oropharyngeal carcinoma (OPC). The assessment was conducted through a retrospective and prospective study, blindly comparing manual plans with deep learning plans. MATERIALS AND METHODS: A set of 95 OPC patients was split into training (n = 60), configuration (n = 10), test retrospective study (n = 10), and test prospective study (n = 15). Our deep learning optimization (DLO) method combines IMPT dose prediction using a deep learning model with a robust mimicking optimization algorithm. Dosimetrists manually adjusted the DLO plan for individual patients. In both studies, manual plans and manually adjusted deep learning (mDLO) plans were blindly assessed by a radiation oncologist, a dosimetrist, and a physicist, through visual inspection, clinical goal evaluation, and comparison of normal tissue complication probability values. mDLO plans were completed within an average time of 2.5 h. In comparison, the manual planning process typically took around 2 days. RESULTS: In the retrospective study, in 10/10 (100%) patients, the mDLO plans were preferred, while in the prospective study, 9 out of 15 (60%) mDLO plans were preferred. In 4 out of the remaining 6 cases, the manual and mDLO plans were considered comparable in quality. Differences between manual and mDLO plans were limited. CONCLUSION: This study showed a high preference for mDLO plans over manual IMPT plans, with 92% of cases considering mDLO plans comparable or superior in quality for OPC patients.

G2PDeep-v2: a web-based deep-learning framework for phenotype prediction and biomarker discovery using multi-omics data.

The G2PDeep-v2 server is a web-based platform powered by deep learning, for phenotype prediction and markers discovery from multi-omics data in any organisms including humans, plants, animals, and viruses. The server provides multiple services for researchers to create deep-learning models through an interactive interface and train these models using an automated hyperparameter tuning algorithm on high-performance computing resources. Users can visualize the results of phenotype and markers predictions and perform Gene Set Enrichment Analysis for the significant markers to provide insights into the molecular mechanisms underlying complex diseases and other biological processes. The G2PDeep-v2 server is publicly available at https://g2pdeep.org/.

MobileNet-V2 /IFHO model for Accurate Detection of early-stage diabetic retinopathy.

Diabetic retinopathy is a serious eye disease that may lead to loss of vision if it is not treated. Early detection is crucial in preventing further vision impairment and enabling timely interventions. Despite notable advancements in AI-based methods for detecting diabetic retinopathy, researchers are still striving to enhance the efficiency of these techniques. Therefore, obtaining an efficient technique in this field is essential. In this research, a new strategy has been proposed to improve the detection of diabetic retinopathy by increasing the accuracy of diagnosis and identifying cases in the initial stages. To achieve this, it has been proposed to integrate the MobileNet-V2 deep learning-based neural network with Improved Fire Hawk Optimizer (IFHO). The MobileNet-V2 network has been renowned for its efficiency and accuracy in image classification tasks, making it a suitable candidate for diabetic retinopathy detection. By combining it with the IFHO, the feature selection process has been optimized, which is essential for identifying relevant patterns and abnormalities related to diabetic retinopathy. The Diabetic Retinopathy 2015 dataset has been used to evaluate the effectiveness of the MobileNet-V2/IFHO model. The study results indicate that the DRMNV2/IFHO model consistently outperforms other methods in terms of precision, accuracy, and recall. Specifically, the model achieves an average precision of 97.521 %, accuracy of 96.986 %, and recall of 98.543 %. Moreover, when compared to advanced techniques, the DRMNV2/IFHO model demonstrates superior performance in specificity, F1-score, and AUC, with average values of 97.233 %, 93.8 %, and 0.927, respectively. These results underscore the potential of the DRMNV2/IFHO model as a valuable tool for improving the accuracy and efficiency of DR diagnosis. Nevertheless, additional validation and testing on larger datasets are required to verify the model's effectiveness and robustness in real-world clinical scenarios.

Bystanders' willingness to assist using automated external defibrillators during cardiac arrest.

The "bystander effect," in which the presence of others inhibits rescue actions, has not been specifically examined in the context of cardiac arrest; understanding this effect in relation to rescue with automated external defibrillators (AEDs) is important. This study aims to identify the presence of others as a factor inhibiting rescue actions using an AED, from a social psychology perspective. We collected data through a web-based questionnaire involving registered residents in all 47 prefectures of Japan. The participants were presented with hypothetical scenarios of witnessing cardiac arrest events at train stations, under sparse or crowded conditions, and with or without the presence of competent parties (e.g., station staff or security guards). Their willingness to intervene was assessed across three levels of rescue behavior: (1) running and calling for help, (2) retrieving an AED, and (3) using an AED. This study found evidence of the bystander effect, indicating that the presence of competent others reduced behavioral interventions by bystanders during out-of-hospital cardiac arrest (OHCA) events. Moreover, the perceived presence of competent parties at the scene of a cardiac arrest reduced bystanders' willingness to initiate rescue under certain circumstances. While many bystanders were willing to initiate rescue efforts in response to calls for help, they resisted rescues involving an AED. This study observes that a bystander effect occurs among bystanders witnessing OHCA, explores the inhibiting effects of identifying competent parties on the initiation of rescue efforts, and suggests that there are significant invisible barriers to using AEDs in rescuing patients with OHCA.

Efficacy and Safety of Different Hybrid Closed Loop Systems for Automated Insulin Delivery in People With Type 1 Diabetes: A Systematic Review and Network Meta-Analysis.

AIMS: To compare the efficacy and safety of different hybrid closed loop (HCL) systems in people with diabetes through a network meta-analysis. METHODS: We searched MEDLINE, EMBASE, CENTRAL and PubMed for randomised clinical trials (RCTs) enrolling children, adolescents and/or adults with type 1 or type 2 diabetes, evaluating Minimed 670G, Minimed 780G, Control-IQ, CamAPS Fx, DBLG-1, DBLHU, and Omnipod 5 HCL systems against other types of insulin therapy, and reporting time in target range (TIR) as outcome. RESULTS: A total of 28 RCTs, all enrolling people with type 1 diabetes, were included. HCL systems significantly increased TIR compared with subcutaneous insulin therapy without continuous glucose monitoring (SIT). Minimed 780G achieved the highest TIR ahead of Control IQ (mean difference (MD) 5.1%, 95% confidence interval (95% CI) [0.68; 9.52], low certainty), Minimed 670G (MD 7.48%, 95% CI [4.27; 10.7], moderate certainty), CamAPS Fx (MD 8.94%, 95% CI [4.35; 13.54], low certainty), and DBLG1 (MD 10.69%, 95% CI [5.73; 15.65], low certainty). All HCL systems decreased time below target range, with DBLG1 (MD -3.69%, 95% CI [-5.2; -2.19], high certainty), Minimed 670G (MD -2.9%, 95% CI [-3.77; -2.04], moderate certainty) and Minimed 780G (MD -2.79%, 95% CI [-3.94; -1.64], high certainty) exhibiting the largest reductions compared to SIT. The risk of severe hypoglycaemia and diabetic ketoacidosis was similar to other types of insulin therapy. CONCLUSIONS: We show a hierarchy of efficacy among the different HCL systems in people with type 1 diabetes, thus providing support to clinical decision-making. TRIAL REGISTRATION: PROSPERO CRD42023453717.

Comparing nursing medication rounds before and after implementation of automated dispensing cabinets: A time and motion study.

Nursing medication administration is an integral, albeit time consuming component of a nursing shift. Automated dispensing cabinets (ADCs) are a medicines management solution designed to improve both efficiency and patient safety. This study aimed to evaluate the time taken to undertake a medication round including the number of locations visited to retrieve medicines, across four different clinical specialties within one hospital. Studies to date have investigated the effect of ADCs on nursing medication rounds centred around one clinical specialty, in hospitals with varying levels of digital maturity. This study adds to the existing body of evidence by investigating multiple clinical specialties where EPMA in use throughout the study period. In this study, prior to ADC implementation nurses retrieved required medicines from shelves in the medication room, mobile medication carts, and patients' own drug (POD) lockers. Post-ADC implementation, medicines were retrieved exclusively from the ADC and POD lockers only. Nurses were observed on each ward completing medication rounds, using the data collection tool designed for this study. Pre-implementation data was collected between February and June 2023, and post-implementation data collected between July and September 2023. There was a statistically significant reduction in the time required for medicines retrieval on the surgical ward only, post- ADC implementation. The time taken to retrieve each medication went from a mean of 98.1 s to 47.2 s (p = 0.0255). When comparing all four specialties as a whole, there was a reduction in the mean time required to issue each medicine preversus post-ADC implementation, from 83.3 s to 62.6 s respectively, however this difference was not shown to be statistically significant. The mean number of locations visited to obtain all required medicines for each patient reduced significantly from 1.73 to 1.04 (p < 0.01). There is potential for improved efficiency as nurses become more familiar with new workflows. It may be of benefit to repeat this study to ascertain whether time savings have been further improved.

Multiview deep learning networks based on automated breast volume scanner images for identifying breast cancer in BI-RADS 4.

Objectives: To develop and validate a deep learning (DL) based automatic segmentation and classification system to classify benign and malignant BI-RADS 4 lesions imaged with ABVS. Methods: From May to December 2020, patients with BI-RADS 4 lesions from Centre 1 and Centre 2 were retrospectively enrolled and divided into a training set (Centre 1) and an independent test set (Centre 2). All included patients underwent an ABVS examination within one week before the biopsy. A two-stage DL framework consisting of an automatic segmentation module and an automatic classification module was developed. The preprocessed ABVS images were input into the segmentation module for BI-RADS 4 lesion segmentation. The classification model was constructed to extract features and output the probability of malignancy. The diagnostic performances among different ABVS views (axial, sagittal, coronal, and multi-view) and DL architectures (Inception-v3, ResNet 50, and MobileNet) were compared. Results: A total of 251 BI-RADS 4 lesions from 216 patients were included (178 in the training set and 73 in the independent test set). The average Dice coefficient, precision, and recall of the segmentation module in the test set were 0.817 ± 0.142, 0.903 ± 0.183, and 0.886 ± 0.187, respectively. The DL model based on multiview ABVS images and Inception-v3 achieved the best performance, with an AUC, sensitivity, specificity, PPV, and NPV of 0.949 (95% CI: 0.945-0.953), 82.14%, 95.56%, 92.00%, and 89.58%, respectively, in the test set. Conclusions: The developed multiview DL model enables automatic segmentation and classification of BI-RADS 4 lesions in ABVS images.

Blocking Air in the Filtering Bleb: Effects on Graft Adhesion After Descemet's Stripping Automated Endothelial Keratoplasty.

BACKGROUND: To evaluate the outcome of Descemet's stripping automated endothelial keratoplasty (DSAEK) with "blocking air" in the filtering bleb in patients with previous trabeculectomy. METHODS: In total, 299 eyes in 283 patients who underwent DSAEK were retrospectively reviewed. Endothelial graft adhesion, intraocular pressure (IOP), and air volume in the anterior chamber with (group A) or without (group B) a filtering bleb were compared between the groups. Group A was divided into two subgroups according to the presence (group A1) or absence (group A2) of air in the filtering bleb; the same three factors were compared between the subgroups. RESULTS: The graft detachment rate was significantly higher in group A (14.3%) than in group B (6.5%) (p = 0.04). IOP was significantly lower in group A than in group B before surgery (p = 0.01), at the end of surgery (p = 0.04), at three hours (p < 0.001), and one week postoperatively (p = 0.02). The graft detachment rate did not significantly differ between groups A1 and A2. There were no differences in IOP at each follow-up time, whereas there was a statistically significant increase in IOP from the preoperative measurement to the end of surgery in group A1 (21.0±7.0 mmHg) compared with group A2 (14.2±8.6 mmHg) (p = 0.02). CONCLUSIONS: The presence of blocking air in the filtering bleb helps ensure increased IOP during the early postoperative period but had no significant effect on graft detachment rates.

Drastic hourly changes in hand hygiene workload and performance rates: A multicenter time series analysis.

BACKGROUND: High hand hygiene (HH) workload is a commonly cited barrier to optimal HH performance. The objective of this study was to assess trends of HH workload as defined by HH opportunities (HHO) and performance rates over different timescales using automated HH monitoring system data. METHODS: This multiyear retrospective observational study was conducted in 58 inpatient units located in 10 North American hospitals. HHO and HH rates were analyzed by time series mixed effects general additive model. RESULTS: Median HH rates peaked at 50.0 between 6 and 7 AM with a trough of 38.2 at 5 PM. HHO over hours in a day were the highest at 184 per hospital unit per hour at 10 AM with a trough of 49.0 between 2 and 3 AM. Median rates for day and night shifts were 40.8 and 45.5, respectively (P = .078). Weekend day shift had the lowest median rate (39.4) compared with any other 12-hour shift (P < .1018). The median rates and HHO varied little across days in a week and months. CONCLUSIONS: HH workload and performance rates were negatively correlated and changed drastically over hours in a day. Hospitals should consider HH workload in the development and timely delivery of improvement interventions.

Advancements in automated classification of chronic obstructive pulmonary disease based on computed tomography imaging features through deep learning approaches.

Chronic Obstructive Pulmonary Disease (COPD) represents a global public health issue that significantly impairs patients' quality of life and overall health. As one of the primary causes of chronic respiratory diseases and global mortality, effective diagnosis and classification of COPD are crucial for clinical management. Pulmonary function tests (PFTs) are standard for diagnosing COPD, yet their accuracy is influenced by patient compliance and other factors, and they struggle to detect early disease pathologies. Furthermore, the complexity of COPD pathological changes poses additional challenges for clinical diagnosis, increasing the difficulty for physicians in practice. Recently, deep learning (DL) technologies have demonstrated significant potential in medical image analysis, particularly for the diagnosis and classification of COPD. By analyzing key radiological features such as airway alterations, emphysema, and vascular characteristics in Computed Tomography (CT) scan images, DL enhances diagnostic accuracy and efficiency, providing more precise treatment plans for COPD patients. This article reviews the latest research advancements in DL methods based on principal radiological features of COPD for its classification and discusses the advantages, challenges, and future research directions of DL in this field, aiming to provide new perspectives for the personalized management and treatment of COPD.