Disease Categories Research Articles

Influence of easing COVID-19 strategies following downgrading of the national infectious disease category on COVID-19 occurrence among hospitalized patients in Japan

PurposeWe aimed to evaluate the influence of easing COVID-19 preventive measures following its downgrading to a common infectious disease on COVID-19 occurrence among hospitalized patients. MethodsNosocomial occurrence of COVID-19 was compared between periods with national infectious disease category 5 versus the preceding category 2 equivalent. Changes in the revised manual included a shorter duration of work restrictions for infected health care practitioners (HCPs); no work restriction for HCPs exposed to SARS-CoV-2 with a negative test on days 1, 3 and 5; discontinuation of universal pre-admission screening; and pre-emptive isolation of patients without screening. Wearing an N95 mask and face shield was required in procedure/care with moderate-to high-risk contact. ResultsAlthough the mean monthly number of infected HCPs increased from 8.1 to 12.7 in the category 5 period (p = 0.034) and that of pre-admission screening decreased to one-fourth, the COVID-19 incidence in hospitalized patients remained similar between the two study periods (1.60 ± 5.59/month versus 1.40 ± 2.63/month, p = 0.358). Clusters, defined as ≥3 COVID-19 patients on the ward, were experienced twice in the preceding period and only once in the category 5 period. The index cases causing nosocomial SARS-CoV-2 transmission mostly involved rehabilitation therapists in the preceding period; five of six index cases were patients in the category 5 period. Following the expanded indication for N95 masks, neither SARS-CoV-2 transmission to patients nor transmission from infected patients was observed in HCPs for 1 year. ConclusionWith sustained, enhanced standard precautions, easing prevention strategies could limit nosocomial SARS-CoV-2 infections.

CDT-CAD: Context-Aware Deformable Transformers for End-to-End Chest Abnormality Detection on X-Ray Images.

Deep learning methods have achieved great success in medical image analysis domain. However, most of them suffer from slow convergency and high computing cost, which prevents their further widely usage in practical scenarios. Moreover, it has been proved that exploring and embedding context knowledge in deep network can significantly improve accuracy. To emphasize these tips, we present CDT-CAD, i.e., context-aware deformable transformers for end-to-end chest abnormality detection on X-Ray images. CDT-CAD first constructs an iterative context-aware feature extractor, which not only enlarges receptive fields to encode multi-scale context information via dilated context encoding blocks, but also captures unique and scalable feature variation patterns in wavelet frequency domain via frequency pooling blocks. Afterwards, a deformable transformer detector on the extracted context features is built to accurately classify disease categories and locate regions, where a small set of key points are sampled, thus leading the detector to focus on informative feature subspace and accelerate convergence speed. Through comparative experiments on Vinbig Chest and Chest Det 10 Datasets, CDT-CAD demonstrates its effectiveness in recognizing chest abnormities and outperforms 1.4% and 6.0% than the existing methods in AP50 and AR on VinBig dateset, and 0.9% and 2.1% on Chest Det-10 dataset, respectively.

Marine Pharmacology in 2019-2021: Marine Compounds with Antibacterial, Antidiabetic, Antifungal, Anti-Inflammatory, Antiprotozoal, Antituberculosis and Antiviral Activities; Affecting the Immune and Nervous Systems, and Other Miscellaneous Mechanisms of Action.

The current 2019-2021 marine pharmacology literature review provides a continuation of previous reviews covering the period 1998 to 2018. Preclinical marine pharmacology research during 2019-2021 was published by researchers in 42 countries and contributed novel mechanism-of-action pharmacology for 171 structurally characterized marine compounds. The peer-reviewed marine natural product pharmacology literature reported antibacterial, antifungal, antiprotozoal, antituberculosis, and antiviral mechanism-of-action studies for 49 compounds, 87 compounds with antidiabetic and anti-inflammatory activities that also affected the immune and nervous system, while another group of 51 compounds demonstrated novel miscellaneous mechanisms of action, which upon further investigation, may contribute to several pharmacological classes. Thus, in 2019-2021, a very active preclinical marine natural product pharmacology pipeline provided novel mechanisms of action as well as new lead chemistry for the clinical marine pharmaceutical pipeline targeting the therapy of several disease categories.

Identifying High-Quality Non-Instrumental Dysphagia Screening Tools for Detection of Adult Dysphagia Case in Acute-Care Settings: A Systematic Review.

As patients nowadays tend to have multiple diseases and complex medical histories, our aim was to identify high-quality, non-instrumental dysphagia screening tools used for the detection of adult dysphagia cases in all disease categories in acute-care settings. A literature search was conducted in five databases from each database's earliest inception to 31 July 2021 and guided by five keywords: 'dysphagia', 'deglutition', 'screening', 'test' and 'measure'. Without limiting the search in any specific disease category, reviewers assessed original studies and identified tools if they had been validated against instrumental evaluations and if they had been designed as a pass-fail procedure to screen whether dysphagia is absent or present. We further excluded any tool if it was (1) for pediatric focus, or (2) a patient self-report questionnaire. All final tool candidates underwent a methodological quality appraisal using the Revised Tool for the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2). Out of 195 studies with 165 tools identified, 20 tool candidates underwent QUADAS-2 review. We found six high-quality, non-instrumental screening tools for detecting adult dysphagia cases in acute-care settings, including the Yale Swallow Protocol, Gugging Swallowing Screen, Toronto Bedside Swallowing Screening Test (both English and Portuguese versions), Sapienza Global Bedside Evaluation of Swallowing and Two-Step Thickened Water Test. These high-quality tools were developed primarily for patients with stroke. Only Yale Swallow Protocol was originally tested for heterogeneous populations with stroke, multiple sclerosis, traumatic brain injury, oesophageal surgery, neurosurgery and head-and-neck cancer. The results highlight the gap in the unavailability of high-quality dysphagia screening tool in several emerged high-risk populations including elderly inpatients, or patients following endotracheal extubation. Further research is needed to determine whether these six tools can be effectively applied across different high-risk populations in acute-care settings to screen for cases finding.

Respiratory Tract Diseases with Musculoskeletal System Interaction: A Scoping Review

Objective: Respiratory diseases and musculoskeletal disorders are significant causes of morbidity and mortality worldwide, especially among older adults and immunocompromised individuals. Although current guidelines encourage a multidimensional approach to diagnosis and treatment, the interaction between these disease categories has not been adequately investigated. This scoping review aims to provide an overview of current research on diseases related to both respiratory and musculoskeletal disorders and identify gaps for future studies. Methods: Using the PRISMA extension to scope the reviews, we focused on randomized controlled trials and cohort studies published in the last five years. A total of 2003 abstracts were identified in databases such as Scopus, WoS, PubMed, Medline, Cinahl, OpenAIRE and EBSCO. After removing 287 duplicates, 1716 articles were screened based on title and abstract. A total of 470 full-text articles were assessed for eligibility, resulting in 11 articles meeting our inclusion criteria. Results: The review shows limited but influential studies investigating the intersection between respiratory and musculoskeletal diseases. The findings suggest that musculoskeletal disorders may negatively affect respiratory functions and vice versa. Several studies have demonstrated potential biomarkers, the importance of sleep quality, and associations with multimorbidity. Conclusion: Although limited to 11 articles, this review highlights the importance of a more detailed understanding of the interactions between the musculoskeletal system and respiratory diseases. This may inform future diagnostic and treatment strategies. However, the limited number of studies in this area indicates that more research is needed, especially which contains interaction mechanism of musculoskeletal and pulmonary pathologies.

Advanced and Metastatic Non-Melanoma Skin Cancer: Epidemiology, Risk Factors, Clinical Features, and Treatment Options.

Non-melanoma skin cancers (NMSC) form the majority of skin cancers, with basal cell carcinoma (BCC) being the most common and cutaneous squamous cell carcinoma (cSCC) being second. Prolonged ultraviolet (UV) exposure, aging, male gender, and immunosuppression represent most of the causes of this category of diseases. BCCs and cSCCs both include different types of skin cancers, such as nodular or morpheaform BCC or flat cSCC. Locally advanced and metastatic NMSCs cannot be treated surgically; thus, systemic therapy (TKI and Immunotherapy) is needed. Interestingly, NMSCs are frequently linked to abnormal Hedgehog (HH) signaling which most systemic immunotherapies for these cancers are based upon. Of note, the first line therapies of BCC, sonidegib and vismodegib, are HH inhibitors. Programmed death receptor 1 antibody (PD-1) inhibitors such as cemiplimab, pembrolizumab, and nivolumab have been approved for the treatment of cSCC. Thus, this paper reviews the epidemiology, risk factors, clinical features, and treatment options for both BCC and cSCC.

MitoPADdb: A database of mitochondrial proteins associated with diseases

Mitochondrial protein/gene mutations and expression variations contribute to the pathogenesis of various diseases, such as neurodegenerative and metabolic diseases. Detailed studies on mitochondrial protein-encoding (MPE) genes across diseases can provide clues for novel therapeutic strategies. Here, we collected, compiled, and manually curated the MPE gene mutation and expression variations data and their association with diseases in a single platform named mitoPADdb. The database contains 810 genes with 18,356 mutations and 1284 qualitative expression variations associated with 1793 diseases, grouped into 15 categories. It allows users to perform a comparative quantitative gene expression analysis for 317 transcriptomic studies across disease categories. Further, it provides information on MPE genes-associated molecular pathways. The mitoPADdb is a valuable resource for investigating mitochondrial dysfunction-related diseases. It can be accessed via http://bicresources.jcbose.ac.in/ssaha4/mitopaddb/index.html.

Ethnomedicinal plants for cardiovascular diseases management in manikganj district

The present research deals with the identification of ethnomedicinal plants used by the local people for curing cardiovascular diseases in Manikganj district, Bangladesh. The present study aimed to record and validate the ethnomedicinal knowledge used by the local people for cardiovascular disease (CVD) management in Manikganj district, Bangladesh using recent ethnobotanical survey methods. The record of 131 ethnomedicinal plant species with diverse use patterns reflected the traditional knowledge richness in the study area. Among the ethnomedicinal plants, 10 plant species, including Phyllanthus emblica L., Allium sativum L., Terminalia arjuna (Roxb. Ex DC.) Wight & Arn, Achyranthes aspera L., Terminalia chebula Retz., Allium cepa L., Adhatoda zeylanica Nees, Cinnamomum tamala Nees & Eberm, Cajanus cajan (L.) Millsp, and Ipomea aquatica Forssk, were cited most frequently by the local people. Among the disease categories, the highest Fic (factor of informant consensus) value was obtained in cardiac complaints, followed by blood purifiers, chest pain, high blood pressure, and antioxidants. In this research, 131 plant species were recorded, and 45 plants obtained the highest Fl (fidelity level) value of 100%. The members of the most common medicinal plants were from the families of Fabaceae, Amaranthaceae, Asteraceae, Cucurbitaceae, and Apocynaceae. This result provides baseline data to link the local population, including traditional health professionals and scientific communities, which could be significant in drug discovery and is very important for the sustainable development program and conservation management in Bangladesh. Bangladesh J. Plant Taxon. 31(1): 101-121, 2024 (June)

Dissecting heritability, environmental risk, and air pollution causal effects using > 50 million individuals in MarketScan

Large national-level electronic health record (EHR) datasets offer new opportunities for disentangling the role of genes and environment through deep phenotype information and approximate pedigree structures. Here we use the approximate geographical locations of patients as a proxy for spatially correlated community-level environmental risk factors. We develop a spatial mixed linear effect (SMILE) model that incorporates both genetics and environmental contribution. We extract EHR and geographical locations from 257,620 nuclear families and compile 1083 disease outcome measurements from the MarketScan dataset. We augment the EHR with publicly available environmental data, including levels of particulate matter 2.5 (PM2.5), nitrogen dioxide (NO2), climate, and sociodemographic data. We refine the estimates of genetic heritability and quantify community-level environmental contributions. We also use wind speed and direction as instrumental variables to assess the causal effects of air pollution. In total, we find PM2.5 or NO2 have statistically significant causal effects on 135 diseases, including respiratory, musculoskeletal, digestive, metabolic, and sleep disorders, where PM2.5 and NO2 tend to affect biologically distinct disease categories. These analyses showcase several robust strategies for jointly modeling genetic and environmental effects on disease risk using large EHR datasets and will benefit upcoming biobank studies in the era of precision medicine.

DNA Methylation and p53 Immunohistochemistry as Prognostic Biomarkers for Vulvar Lichen Sclerosus

Vulvar lichen sclerosus (LS) is an inflammatory dermatosis that can progress to human papillomavirus (HPV)-independent vulvar intraepithelial neoplasia (HPVi VIN) and vulvar squamous cell carcinoma (VSCC). Although LS has a much lower cancer risk compared with HPVi VIN (5% vs 50%, respectively), its incidence is significantly higher. Therefore, there is a clinical need to identify LS patients with an increased cancer risk. Our objective was to study the value of DNA methylation and p53 immunohistochemistry (IHC) as prognostic biomarkers for progression to cancer in patients with LS. Vulvar tissues from 236 patients were selected, including 75 LS and 68 HPVi VIN, both with and without cancer development, 32 VSCC, and 61 healthy vulvar controls. Samples were subjected to p53 IHC and DNA methylation analysis of a 3-gene marker panel containing ZNF582, SST, and miR124-2. Methylation levels and p53 IHC status (mutant or wild-type) were assessed and compared among all disease categories. Odds ratios were determined to identify whether the biomarkers were associated with progression to cancer in patients with LS. The highest methylation levels were found in HPVi VIN and VSCC, followed by LS and healthy vulvar controls. The largest heterogeneity in methylation levels was observed in LS cases. In fact, the 3-marker panel tested positive in 70% of LS, which progressed to VSCC vs only 17% of LS in patients without cancer development (P = .002). Also, mutant p53 IHC was observed more frequently in LS with progression to VSCC compared with nonprogressive LS cases (42% vs 3%, respectively, P = .001). Multivariable analysis identified a mutant p53 status as the only independent risk factor for cancer development in LS (odds ratio: 34.0, 95% CI: 1.4-807.4). In conclusion, DNA methylation testing and p53 IHC show strong potential as prognostic biomarkers for the identification of LS patients at high risk of progression to cancer.

Topically Applied Therapies for the Treatment of Skin Disease: Past, Present, and Future.

The purpose of this review is to summarize essential biological, pharmaceutical, and clinical aspects in the field of topically applied medicines that may help scientists when trying to develop new topical medicines. After a brief history of topical drug delivery, a review of the structure and function of the skin and routes of drug absorption and their limitations is provided. The most prevalent diseases and current topical treatment approaches are then detailed, the organization of which reflects the key disease categories of autoimmune and inflammatory diseases, microbial infections, skin cancers, and genetic skin diseases. The complexity of topical product development through to large-scale manufacturing along with recommended risk mitigation approaches are then highlighted. As such topical treatments are applied externally, patient preferences along with the challenges they invoke are then described, and finally the future of this field of drug delivery is discussed, with an emphasis on areas that are more likely to yield significant improvements over the topical medicines in current use or would expand the range of medicines and diseases treatable by this route of administration. SIGNIFICANCE STATEMENT: This review of the key aspects of the skin and its associated diseases and current treatments along with the intricacies of topical formulation development should be helpful in making judicious decisions about the development of new or improved topical medicines. These aspects include the choices of the active ingredients, formulations, the target patient population's preferences, limitations, and the future with regard to new skin diseases and topical medicine approaches.

Changes in the prevalence of maternal chronic conditions during pregnancy: A nationwide age-period-cohort analysis.

To estimate temporal changes in the prevalence of pre-existing chronic conditions among pregnant women in Sweden and evaluate the extent to which secular changes in maternal age, birth cohorts and obesity are associated with these trends. Population-based cross-sectional study. Sweden, 2002-2019. All women (aged 15-49 years) who delivered in Sweden (2002-2019). An age-period-cohort analysis was used to evaluate the effects of age, calendar periods, and birth cohorts on the observed temporal trends. Pre-existing chronic conditions, including 17 disease categories of physical and psychiatric health conditions recorded within 5 years before childbirth, presented as prevalence rates and rate ratios (RRs) with 95% confidence intervals (CIs). Temporal trends were also adjusted for pre-pregnancy body mass index (BMI) and the mother's country of birth. The overall prevalence of at least one pre-existing chronic condition was 8.7% (147 458 of 1 703 731 women). The rates of pre-existing chronic conditions in pregnancy increased threefold between 2002-2006 and 2016-2019 (RR 2.82, 95% CI 2.77-2.87). Rates of psychiatric (RR 3.80, 95% CI 3.71-3.89), circulatory/metabolic (RR 1.62, 95% CI 1.55-1.71), autoimmune/neurological (RR 1.69, 95% CI 1.61-1.78) and other (RR 2.10, 95% CI 1.99-2.22) conditions increased substantially from 2002-2006 to 2016-2019. However, these increasing rates were less pronounced between 2012-2015 and 2016-2019. No birth cohort effect was evident for any of the pre-existing chronic conditions. Adjusting for secular changes in obesity and the mother's country of birth did not affect these associations. The burden of pre-existing chronic conditions in pregnancy in Sweden increased from 2002 to 2019. This increase may be associated with the improved reporting of diagnoses and advancements in chronic condition treatment among women, potentially enhancing their fecundity.

Between-hospital variation in indicators of quality of care: a systematic review

BackgroundEfforts to mitigate unwarranted variation in the quality of care require insight into the ‘level’ (eg, patient, physician, ward, hospital) at which observed variation exists. This systematic literature review aims...

A comparative evaluation of ChatGPT 3.5 and ChatGPT 4 in responses to selected genetics questions.

To evaluate the efficacy of ChatGPT 4 (GPT-4) in delivering genetic information about BRCA1, HFE, and MLH1, building on previous findings with ChatGPT 3.5 (GPT-3.5). To focus on assessing the utility, limitations, and ethical implications of using ChatGPT in medical settings. A structured survey was developed to assess GPT-4's clinical value. An expert panel of genetic counselors and clinical geneticists evaluated GPT-4's responses to these questions. We also performed comparative analysis with GPT-3.5, utilizing descriptive statistics and using Prism 9 for data analysis. The findings indicate improved accuracy in GPT-4 over GPT-3.5 (P < .0001). However, notable errors in accuracy remained. The relevance of responses varied in GPT-4, but was generally favorable, with a mean in the "somewhat agree" range. There was no difference in performance by disease category. The 7-question subset of the Bot Usability Scale(BUS-15) showed no statistically significant difference between the groups but trended lower in the GPT-4 version. The study underscores GPT-4's potential role in genetic education, showing notable progress yet facing challenges like outdated information and the necessity of ongoing refinement. Our results, while showing promise, emphasizes the importance of balancing technological innovation with ethical responsibility in healthcare information delivery.

Vision Transformer Based Classification of Neurological Disorders from Human Speech

In this study, we introduce a transformative approach to achieve high-accuracy classification of distinct health categories, including Parkinson's disease, Multiple Sclerosis (MS), healthy individuals, and other categories, utilizing a transformer-based neural network. The cornerstone of this approach lies in the innovative conversion of human speech into spectrograms, which are subsequently transformed into visual images. This transformation process enables our network to capture intricate vocal patterns and subtle nuances that are indicative of various health conditions. The experimental validation of our approach underscores its remarkable performance, achieving exceptional accuracy in differentiating Parkinson's disease, MS, healthy subjects, and other categories. This breakthrough opens doors to potential clinical applications, offering an innovative, non-invasive diagnostic tool that rests on the fusion of spectrogram analysis and transformer-based models.

Disease detection and classification in pomegranate fruit using hybrid convolutional neural network with honey badger optimization algorithm

ABSTRACT Fruit cultivation plays a pivotal role in improvement of the agricultural economy. Pomegranate is a nutritionally rich fruit that is highly valuable because of its excellent antioxidant properties, richness in vitamins and fiber. Pomegranate is affected by diseases, which can severely affect its yield and quality. Timely disease detection is vital for effective disease control in pomegranate orchards. This study presents a novel approach for disease detection in pomegranate and categorization of pomegranate using a hybrid Convolutional Neural Network (CNN) coupled with the Honey Badger Optimization Algorithm (HBOA). The proposed hybrid CNN-HBOA approach leverages the strengths of both CNN and metaheuristic optimization algorithm HBOA. The CNN model was designed to extract high-level features from images of pomegranate affected by various diseases, and HBOA was employed to fine-tune the CNN parameters to improve the accuracy and robustness of the model. Initially, the quality of the images was enhanced using a contrast enhancement technique, and the images were segmented using the Detectron2 algorithm. Finally, the extracted features were fed as input to the disease detection stage wherein the CNN classifies the disease. Furthermore, the classification accuracy of the CNN model was enhanced using the HBOA algorithm. The performance of the hybrid CNN-HBOA algorithm was evaluated using a vast collection of images depicting multiple disease categories of pomegranate. Experimental results demonstrate the superiority of the hybrid CNN-HBOA system, which achieves a detection accuracy of 99.58%, precision of 100%, recall of 99.71% and F1 score of 99.75% compared to other existing state-of-the-art models.

Thinking Beyond Disease Silos: Dysregulated Genes Common in Tuberculosis and Lung Cancer as Identified by Systems Biology and Machine Learning.

The traditional way of thinking about human diseases across clinical and narrow phenomics silos often masks the underlying shared molecular substrates across human diseases. One Health and planetary health fields particularly address such complexities and invite us to think across the conventional disease nosologies. For example, tuberculosis (TB) and lung cancer (LC) are major pulmonary diseases with significant planetary health implications. Despite distinct etiologies, they can coexist in a given community or patient. This is both a challenge and an opportunity for preventive medicine, diagnostics, and therapeutics innovation. This study reports a bioinformatics analysis of publicly available gene expression data, identifying overlapping dysregulated genes, downstream regulators, and pathways in TB and LC. Analysis of NCBI-GEO datasets (GSE83456 and GSE103888) unveiled differential expression of CEACAM6, MUC1, ADM, DYSF, PLOD2, and GAS6 genes in both diseases, with pathway analysis indicating association with lysine degradation pathway. Random forest, a machine-learning-based classification, achieved accuracies of 84% for distinguishing TB from controls and 83% for discriminating LC from controls using these specific genes. Additionally, potential drug targets were identified, with molecular docking confirming the binding affinity of warfarin to GAS6. Taken together, the present study speaks of the pressing need to rethink clinical diagnostic categories of human diseases and that TB and LC might potentially share molecular substrates. Going forward, planetary health and One Health scholarship are poised to cultivate new ways of thinking about diseases not only across medicine and ecology but also across traditional diagnostic conventions.

Advanced Deep Learning Models for Accurate Retinal Disease State Detection

Retinal diseases are a significant challenge in the realm of medical diagnosis, with potential complications to vision and overall ocular health. This research endeavors to address the challenge of automating the detection of retinal disease states using advanced deep learning models, including VGG-19, ResNet-50, InceptionV3, and EfficientNetV2. Each model leverages transfer learning, drawing insights from a substantial dataset comprising optical coherence tomography (OCT) images and subsequently classifying images into four distinct retinal conditions: choroidal neovascularization, drusen, diabetic macular edema and a healthy state. The training dataset, sourced from repositories that are available to the public including OCT retinal images, spanning all four disease categories. Our findings reveal that among the models tested, EfficientNetV2 demonstrates superior performance, with a remarkable classification accuracy of 98.92%, precision of 99.6%, and a recall of 99.4%, surpassing the performance of the other models.

"If I have a limited amount of time left, this is not how I want to spend it": A qualitative descriptive study of factors influencing daily life participation following a cancer diagnosis.

Following a cancer diagnosis, restricted participation in daily life is common. Restricted participation can be temporary or long lasting. The aim of this study was to characterize how daily life participation is impacted following a cancer diagnosis. Eligible individuals included adults (>18years) with any stage/grade brain, breast, colorectal, or lung cancer in any phase of treatment or post-treatment. Participants completed a semi-structured interview about how their life participation was impacted following their cancer diagnosis. Data were analyzed through team-based thematic analysis. Forty adults, 10 per disease category, participated. Four themes were identified that supported or hindered daily life participation: (1) self-expectations, (2) expectations of others, (3) awareness of mortality, and (4) symptoms and side effects of cancer. Participants discussed how their cancer experience resulted in a reprioritization of what they valued doing in their life. However, many survivors struggled to adapt and described a tension between their need to adapt to their current life circumstances and their contrasting desire to stay connected with their pre-cancer selves through daily life participation. The mental health challenges associated with decreased participation were also outlined by participants. Cancer survivors' daily life participation is influenced by expectations from themselves and others, awareness of mortality, and disease symptoms/side effects. Future interventions can target these domains to supports survivors' life participation.

Improved Tomato Disease Detection with YOLOv5 and YOLOv8

This study delves into the application of deep learning for precise tomato disease detection, focusing on four crucial categories: healthy, blossom end rot, splitting rotation, and sun-scaled rotation. The performance of two lightweight object detection models, namely YOLOv5l and YOLOv8l, was compared on a custom tomato disease dataset. Initially, both models were trained without data augmentation to establish a baseline. Subsequently, diverse data augmentation techniques were obtained from Roboflow to significantly expand and enrich the dataset content. These techniques aimed to enhance the models' robustness to variations in lighting, pose, and background conditions. Following data augmentation, the YOLOv5l and YOLOv8l models were re-trained and their performance across all disease categories was meticulously analyzed. After data augmentation, a significant improvement in accuracy was observed for both models, highlighting its effectiveness in bolstering the models' ability to accurately detect tomato diseases. YOLOv8l consistently achieved slightly higher accuracy compared to YOLOv5l, particularly when excluding background images from the evaluation.