Tuberculosis (TB), an infectious disease caused by Mycobacterium tuberculosis, continues to be a major global health threat despite being preventable and curable. This burden is particularly high in low and middle income countries. Microscopy remains essential for diagnosing TB by enabling direct visualization of Mycobacterium tuberculosis in sputum smear samples, offering a cost-effective approach for early detection and effective treatment. Given the labor-intensive nature of microscopy, automating the detection of bacilli in microscopic images is crucial to improve both the expediency and reliability of TB diagnosis. The current methodologies for detecting tuberculosis bacilli in bright field microscopic sputum smear images are hindered by limited automation capabilities, inconsistent segmentation quality, and constrained classification precision. This paper proposes a two-stage deep learning methodology for tuberculosis bacilli detection, comprising bacilli segmentation followed by classification. In the initial phase, an advanced U-Net model employing attention blocks and residual connections is proposed to segment microscopic sputum smear images, enabling the extraction of Regions of Interest (ROIs). The extracted ROIs are then classified using a Vision Transformer, which we specifically customized as TBViT to enhance the precise detection of bacilli within the images. For the experiments, a newly developed dataset of microscopic sputum smear images derived from Ziehl-Neelsen-stained slides is used in conjunction with existing public datasets. The qualitative and quantitative evaluation of the experiments using various metrics demonstrates that the proposed model achieves significantly improved segmentation performance, higher classification accuracy, and a greater level of automation, surpassing existing methods.

The microscopic detection of Mycobacterium tuberculosis bacilli in Ziehl-Neelsen (ZN)-stained sputum smears remains essential for tuberculosis (TB) diagnosis, but it is limited by inter-observer variability, high operational workload, and a shortage of qualified specialists. Deep learning (DL) has shown strong potential to support this task; however, progress is constrained by the limited availability of well-annotated datasets and the lack of standardized evaluation protocols. This study presents a comprehensive investigation of DL-based object detection methods for TB bacillus identification, comprising (i) a systematic comparison of seven detectors, (ii) the introduction of a new annotated ZN-stained dataset containing 500 images, and (iii) an image-partitioning strategy designed to mitigate the performance degradation caused by heterogeneous image resolutions. Three public datasets and the proposed dataset were evaluated using cross-validation and cross-dataset experiments. Among the seven methods tested, Faster R-CNN consistently achieved the highest overall precision, sensitivity, F1-score, and average precision across datasets. The proposed partitioning strategy significantly improved detection performance, particularly for large images. Cross-dataset experiments revealed limited generalization across laboratories, underscoring the importance of standardized datasets and context-aware training. Overall, the proposed dataset and experimental findings offer practical guidance for future research toward robust, clinically deployable AI-assisted TB diagnostic workflows.

Cryptosporidium spp. are significant enteric protozoan parasites affecting neonatal calves, and they are known to potentially have zoonotic implications. An accurate diagnosis is essential for effective disease control and surveillance. This cross-sectional study comparatively evaluated five diagnostic techniques; direct wet smear, Leishman staining, modified Ziehl-Neelsen (mZN) staining, phenolic auramine staining, and polymerase chain reaction (PCR) for detection of Cryptosporidium spp. in diarrheic bovine calves from western Uttar Pradesh, India. During June 2003 to May 2004, among 200 faecal samples from calves aged ≤3 months, 33 positive samples were identified using PCR targeting the COWP gene yielding a molecular prevalence of 16.5%. Using PCR as the reference standard, the sensitivities of wet smear, Leishman, mZN, and auramine were 60.6% (wet smear), 78.8% (Leishman), 90.9% (mZN), and 97.0% (auramine), while the specificities ranged from 94.0% to 99.4%. Auramine staining demonstrated almost perfect agreement with PCR (κ = 0.96), followed by mZN staining (κ = 0.94). These findings confirm the endemic occurrence of Cryptosporidium infection in young calves in this region. Phenolic auramine staining provides a highly reliable and practical alternative to PCR under regional laboratory conditions.

Cryptosporidium infection is a well-established, widely recognized pathogen affecting neonatal dairy calves worldwide, with profound implications for both animal welfare and public health. This investigation evaluated the prevalence, molecular characteristics, and risk determinants of Cryptosporidium infection across seven districts of the Central Plain Zone (CPZ) of Punjab, India. A total of 341 fecal samples from dairy calves (180 buffalo and 161 cattle) were analyzed using modified Ziehl-Neelsen (mZN) staining of direct fecal smears (DFS) and saturated sugar flotation smears (SSFS), the latter employed to enhance oocyst concentration and improve detection sensitivity. Microscopic examination revealed an overall prevalence of 23.75% (81/341), with 39 samples testing positive by DFS and 81 by SSFS, establishing SSFS as the superior diagnostic method. Geographic analysis demonstrated significant spatial heterogeneity, with Ludhiana district exhibiting the highest prevalence (38.46%, P < 0.05) and Shri Fatehgarh Sahib the lowest (6.25%). Multifactorial risk analysis identified four critical determinants of infection: periurban location (38.46%, P < 0.01), age < 1 month (32.4%, P < 0.01), diarrheic status (29.59%, P < 0.01), and organized farm management systems (28.72%, P < 0.05). Molecular characterization through nested PCR amplification of the 18S rRNA gene (∼834 bp) followed by PCR-RFLP analysis using restriction enzymes (SspI, VspI, and MboII) conclusively identified Cryptosporidium parvum as the sole species present in all positive samples from both cattle and buffalo calves. BLASTn analysis showed 98.86% and 98.61% homology with GenBank C. parvum sequences for cattle and buffalo isolates, respectively. Phylogenetic analysis revealed that Punjab isolates clustered with C. parvum strains from diverse geographic regions, including other Indian states (Uttar Pradesh, Bihar) and international locations (Algeria, Bangladesh, China), while remaining distinct from C. bovis, C. ryanae, C. andersoni, and C. hominis. The exclusive detection of C. parvum, based on the few samples analysis, needs further gp60 subtyping to delineate circulating IIa genotypes on more numbers of samples in remaining zones of the Punjab for exploring the zoonotic potential as risk posed by infected dairy calves and underscores the urgent need for comprehensive surveillance and control measures accordingly.

Objectives: Cryptosporidium is a protozoan parasite that commonly infects livestock, including Capra hircus, causing diarrhea, dehydration, and sometimes mortality. These infections lead to substantial economic losses in the farming industry. This study aimed to determine the prevalence and associated risk factors, such as region, season, age, and sex, of Cryptosporidium infection in C. hircus in southern Khyber Pakhtunkhwa, Pakistan. Materials and Methods: A total of 360 fecal samples were collected from C. hircus from the Kohat, Bannu, and Lakki Marwat Districts of Khyber Pakhtunkhwa, Pakistan. Fecal smears were stained using the modified Ziehl-Neelsen method, and statistical analysis, including chi-square tests, was employed to assess the association between Cryptosporidium infection prevalence and various factors. Results: The overall Cryptosporidium infection prevalence was 20.55% (74/360). District-wise analysis revealed the highest prevalence in Kohat (23.33%), followed by Bannu (20%) and Lakki Marwat (18.33%). Monthly examination identified August as the peak prevalence month (36.67%), displaying non-significant monthly variations (p = 0.075). Seasonal disparity indicated a significantly higher prevalence in the summer (30%) compared to other seasons (p = 0.008). Age-wise prevalence demonstrated susceptibility in kids aged ≤ 15 days (≤ 33.93%), with significant overall age-based differences (p = 0.001). Female kids exhibited a slightly higher prevalence at 22.22% compared to males (19.19%), with no significant gender-based difference (p = 0.479). Conclusions: This study highlights the influence of geographic, seasonal, and demographic factors on Cryptosporidium prevalence in goat populations. The findings emphasize the importance of implementing targeted preventive measures. Recommended strategies include farmer education programs, seasonal deworming schedules, early screening for infection, and strict hygiene and water sanitation practices to reduce transmission risks effectively.

Non-tuberculous mycobacteria are recognized as pathogens affecting aquatic animals and humans. Although Mycobacterium marinum is traditionally considered the principal etiological agent of fish mycobacteriosis and fish-tank granuloma in humans, the contribution of Mycobacterium chelonae in fish as well as in humans remains underestimated, particularly in Ziehl-Neelsen (ZN)-negative lesions. In this study, granulomatous lesions of marine fish species and human skin biopsies were investigated using histopathology and molecular diagnostics. Following histological evaluation, all samples were analysed by PCR amplification and sequencing of the hsp65 gene to identify the causative agent. Fish and human tissues showed similar chronic granulomatous lesions, with all cases negative for acid-fast bacilli by ZN staining. Molecular analysis consistently identified M. chelonae in both marine fish organs and human skin lesions exhibiting clinicopathological features traditionally attributed to M. marinum. Phylogenetic analysis revealed high genetic similarity between fish- and human-derived sequences, suggesting shared environmental reservoirs rather than host-specific lineages. These findings identify M. chelonae as an underrecognized cause of granulomatous disease in fish and humans and demonstrate the limitations of histochemical methods in these infections. Incorporating molecular diagnostics into routine investigations of granulomatous disease supports more accurate diagnosis across veterinary and human medicine and strengthens One Health surveillance of aquatic non-tuberculous mycobacterial infections.

Nontuberculous mycobacteria (NTM) represent an increasingly significant cause of pulmonary and extrapulmonary infections, but are sometimes misinterpreted as tuberculosis (TB) owing to overlapping clinical and microbiological characteristics. Conventional diagnostic approaches, such as Ziehl-Neelsen staining and culture in a Mycobacterial Growth Indicator Tube (MGIT) system, are constrained by extended incubation times, are insufficient for accurate species differentiation, and are limited by prolonged incubation periods. Recent molecular and genomic advances have transformed NTM diagnostics by enabling rapid, specific, and high-resolution identification. Line probe assays (e.g., GenoType Mycobacterium CM/AS assay) and multiplex PCR have enhanced the ability to distinguish between NTM species such as Mycobacterium absessus, M. fortuitum, and M. avium complex and M. tuberculosis complex, which is essential for proper treatment and epidemiological mapping. Among newer proteomic platforms, matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry has emerged as a transformative, cost-effective technology capable of identifying Mycobacterium species directly from culture isolates through protein fingerprinting. It provides rapid, reproducible, and highly discriminatory identification between closely related species. Next-generation sequencing (NGS) and whole genome sequencing approaches now offer unprecedented insight into species identification, strain typing, and drug-resistance prediction, complementing traditional culture-based susceptibility testing. Newer techniques such as metagenomics NGS (mNGS), targeted NGS (tNGS) multilocus sequence typing, and mycobacterial interspersed repetitive unit-variable number tandem repeats (MIRU-VNTR) genotyping facilitate subspecies-level resolution and real-time outbreak surveillance. Moreover, molecular beacons, insertion sequence analysis, and repetitive sequence-based polymerase chain reaction (Rep-PCR) enhance detection sensitivity even in paucibacillary samples. The integration of genomic data with automated diagnostic system promises earlier intervention, accurate species delineation, and improved patient outcome.