Analysis of banana plant health using machine learning techniques.

The Indian economy is greatly influenced by the Banana Industry, necessitating advancements in agricultural farming. Recent research emphasizes the imperative nature of addressing diseases that impact Banana Plants, with a particular focus on early detection to safeguard production. The urgency of early identification is underscored by the fact that diseases predominantly affect banana plant leaves. Automated systems that integrate machine learning and deep learning algorithms have proven to be effective in predicting diseases. This manuscript examines the prediction and detection of diseases in banana leaves, exploring various diseases, machine learning algorithms, and methodologies. The study makes a contribution by proposing two approaches for improved performance and suggesting future research directions. In summary, the objective is to advance understanding and stimulate progress in the prediction and detection of diseases in banana leaves. The need for enhanced disease identification processes is highlighted by the results of the survey. Existing models face a challenge due to their lack of rotation and scale invariance. While algorithms such as random forest and decision trees are less affected, initially convolutional neural networks (CNNs) is considered for disease prediction. Though the Convolutional Neural Network models demonstrated impressive accuracy in many research but it lacks in invariance to scale and rotation. Moreover, it is observed that due its inherent design it cannot be combined with feature extraction methods to identify the banana leaf diseases. Due to this reason two alternative models that combine ANN with scale-invariant Feature transform (SIFT) model or histogram of oriented gradients (HOG) combined with local binary patterns (LBP) model are suggested. The first model ANN with SIFT identify the disease by using the activation functions to process the features extracted by the SIFT by distinguishing the complex patterns. The second integrate the combined features of HOG and LBP to identify the disease thus by representing the local pattern and gradients in an image. This paves a way for the ANN to learn and identify the banana leaf disease. Moving forward, exploring datasets in video formats for disease detection in banana leaves through tailored machine learning algorithms presents a promising avenue for research.

State-of-art technologies, challenges, and emerging trends of computer vision in dental images.

Computer vision falls under the broad umbrella of artificial intelligence that mimics human vision and plays a vital role in dental imaging. Dental practitioners visualize and interpret teeth, and the structure surrounding the teeth and detect abnormalities by manually examining various dental imaging modalities. Due to the complexity and cognitive difficulty of comprehending medical data, human error makes correct diagnosis difficult. Automated diagnosis may be able to help alleviate delays, hasten practitioners' interpretation of positive cases, and lighten their workload. Several medical imaging modalities like X-rays, CT scans, color images, etc. that are employed in dentistry are briefly described in this survey. Dentists employ dental imaging as a diagnostic tool in several specialties, including orthodontics, endodontics, periodontics, etc. In the discipline of dentistry, computer vision has progressed from classic image processing to machine learning with mathematical approaches and robust deep learning techniques. Here conventional image processing techniques solely as well as in conjunction with intelligent machine learning algorithms, and sophisticated architectures of dental radiograph analysis employ deep learning techniques. This study provides a detailed summary of several tasks, including anatomical segmentation, identification, and categorization of different dental anomalies with their shortfalls as well as future perspectives in this field.

Prevalence of co-occurring diagnoses in people exposed to alcohol prenatally: Findings from a meta-analysis.

BACKGROUND: Individuals with prenatal alcohol exposure (PAE) commonly experience co-occurring diagnoses, which are often overlooked and misdiagnosed and have detrimental impacts on accessing appropriate services. The prevalence of these co-occurring diagnoses varies widely in the existing literature and has not been examined in PAE without an FASD diagnosis. METHOD: A search was conducted in five databases and the reference sections of three review papers, finding a total of 2180 studies. 57 studies were included in the final analysis with a cumulative sample size of 29,644. Bayesian modeling was used to determine aggregate prevalence rates of co-occurring disorders and analyze potential moderators. RESULTS: 82 % of people with PAE had a co-occurring diagnosis. All disorders had a higher prevalence in individuals with PAE than the general population with attention deficit hyperactivity disorder, learning disorder, and intellectual disability (ID) being the most prevalent. Age, diagnostic status, and sex moderated the prevalence of multiple disorders. LIMITATIONS: While prevalence of disorders is crucial information, it does not provide a direct representation of daily functioning and available supports. Results should be interpreted in collaboration with more individualized research to provide the most comprehensive representation of the experience of individuals with PAE. CONCLUSIONS: Co-occurring diagnoses are extremely prevalent in people with PAE, with older individuals, females, and those diagnosed with FASD being most at risk for having a co-occurring disorder. These findings provide a more rigorous examination of the challenges faced by individuals with PAE than has existed in the literature, providing clinicians with information to ensure early identification and effective treatment of concerns to prevent lifelong challenges.

Designing a deep hybridized residual and SE model for MRI image-based brain tumor prediction.

Deep learning techniques have become crucial in the detection of brain tumors but classifying numerous images is time-consuming and error-prone, impacting timely diagnosis. This can hinder the effectiveness of these techniques in detecting brain tumors in a timely manner. To address this limitation, this study introduces a novel brain tumor detection system. The main objective is to overcome the challenges associated with acquiring a large and well-classified dataset. The proposed approach involves generating synthetic Magnetic Resonance Imaging (MRI) images that mimic the patterns commonly found in brain MRI images. The system utilizes a dataset consisting of small images that are unbalanced in terms of class distribution. To enhance the accuracy of tumor detection, two deep learning models are employed. Using a hybrid ResNet+SE model, we capture feature distributions within unbalanced classes, creating a more balanced dataset. The second model, a tailored classifier identifies brain tumors in MRI images. The proposed method has shown promising results, achieving a high detection accuracy of 98.79%. This highlights the potential of the model as an efficient and cost-effective system for brain tumor detection.

Unlocking the Power of Influenza Vaccines for Pediatric Population: A Narrative Review.

The flu, often known as influenza, is a dangerous public health hazard for the pediatric population. Immunization is essential for decreasing the burden of the disease and avoiding complications related to influenza. However, the immunogenicity, efficacy, and safety of different influenza vaccines in children warrant careful evaluation. The purpose of this narrative review is to give a summary of the existing literature on the immunogenicity, efficacy, and safety of several vaccinations against influenza viruses in children. The review incorporates evidence from a range of studies focusing on the outcomes of interest. Immunogenicity studies have shown that influenza vaccines induce a robust immune response in children, primarily through neutralizing antibodies' formation. However, variations in vaccine composition influence the duration and magnitude of immune responses. Safety is a crucial consideration in pediatric vaccination. In children, influenza vaccinations have generally shown a high safety profile, with mild and temporary side effects being the most common. Vaccinations against influenza have shown a modest level of efficacy in avoiding hospitalizations linked to influenza, laboratory-confirmed influenza infections, and serious consequences in children. Live attenuated vaccines have shown higher effectiveness against matched strains compared to inactivated vaccines. In conclusion, this narrative review highlights that receiving influenza vaccination in children aged six to 47 months is very important. While different vaccines exhibit varying immunogenicity, safety profiles, and effectiveness, they all contribute to reducing the burden of influenza among children. Future research should focus on optimizing vaccine strategies, improving vaccine coverage, and evaluating long-term protection.

A Comprehensive Review on Machine Learning Techniques for Protein Family Prediction.

Proteomics is a field dedicated to the analysis of proteins in cells, tissues, and organisms, aiming to gain insights into their structures, functions, and interactions. A crucial aspect within proteomics is protein family prediction, which involves identifying evolutionary relationships between proteins by examining similarities in their sequences or structures. This approach holds great potential for applications such as drug discovery and functional annotation of genomes. However, current methods for protein family prediction have certain limitations, including limited accuracy, high false positive rates, and challenges in handling large datasets. Some methods also rely on homologous sequences or protein structures, which introduce biases and restrict their applicability to specific protein families or structures. To overcome these limitations, researchers have turned to machine learning (ML) approaches that can identify connections between protein features and simplify complex high-dimensional datasets. This paper presents a comprehensive survey of articles that employ various ML techniques for predicting protein families. The primary objective is to explore and improve ML techniques specifically for protein family prediction, thus advancing future research in the field. Through qualitative and quantitative analyses of ML techniques, it is evident that multiple methods utilizing a range of classifiers have been applied for protein family prediction. However, there has been limited focus on developing novel classifiers for protein family classification, highlighting the urgent need for improved approaches in this area. By addressing these challenges, this research aims to enhance the accuracy and effectiveness of protein family prediction, ultimately facilitating advancements in proteomics and its diverse applications.

Sex differences in surgical site infections following coronary artery bypass grafting: a retrospective observational study.

BACKGROUND: Surgical site infection (SSI) following cardiac surgery poses a significant challenge for healthcare providers. Despite advances in surgical techniques and infection control measures, SSI remains a leading cause of morbidity and mortality, in addition to being a significant economic burden on healthcare services. Current literature suggests there is a reproducible difference in the incidence of SSI following cardiac surgery between sexes. We aim to assess the sex-specific predictive risk factors for sternal SSI following coronary artery bypass grafting (CABG) in addition to identifying any differences in the causative organisms between groups. METHODS: Adult patients undergoing isolated CABG between January 2012 and December 2022 in one UK hospital organization were included. In this 10-year, retrospective observational study, a total of 10,208 patients met the inclusion criteria. Pre-operative risk factors were identified using univariate analysis. To assess dependence between sex and organism or Gram stain, a Pearson Chi-squared test with Yates correction for continuity was performed. RESULTS: In total there were 8457 males of which 181 developed a sternal SSI (2.14%) and 1751 females, 128 of whom had a sternal SSI (7.31%). Male patients were found to be significantly more likely to develop an SSI secondary to a Gram-positive organism, whereas female patients were more likely to have a Gram-negative causative organism (P<0.00001). Staphylococcus was statistically more likely to be the causative organism genus in male patients. Pseudomonas aeruginosa was found to be twice as common in the female cohort compared with the male group. CONCLUSION: In our study, we found a statistically significant difference in the causative organisms and Gram stain for post-CABG sternal SSIs between males and females. Male patients predominately have Gram-positive associated SSIs, whereas female SSI pathogens are more likely to be Gram negative. The preoperative risk profiles of both cohorts are similar, including being an insulin-dependent diabetic and triple vessel coronary artery disease. Given these findings, it prompts the question, should we be tailoring our SSI treatment strategies according to sex and associated risk profiles?

Performance optimization of energy-efficient solar absorbers for thermal energy harvesting in modern industrial environments using a solar deep learning model.

Thermal energy harvesting has seen a rise in popularity in recent years due to its potential to generate renewable energy from the sun. One of the key components of this process is the solar absorber, which is responsible for converting solar radiation into thermal energy. In this paper, a smart performance optimization of energy efficient solar absorber for thermal energy harvesting is proposed for modern industrial environments using solar deep learning model. In this model, data is collected from multiple sensors over time that measure various environmental factors such as temperature, humidity, wind speed, atmospheric pressure, and solar radiation. This data is then used to train a machine learning algorithm to make predictions on how much thermal energy can be harvested from a particular panel or system. In a computational range, the proposed solar deep learning model (SDLM) reached 83.22 % of testing and 91.72 % of training results of false positive absorption rate, 69.88 % of testing and 81.48 % of training results of false absorption discovery rate, 81.40 % of testing and 72.08 % of training results of false absorption omission rate, 75.04 % of testing and 73.19 % of training results of absorbance prevalence threshold, and 90.81 % of testing and 78.09 % of training results of critical success index. The model also incorporates components such as insulation and orientation to further improve its accuracy in predicting the amount of thermal energy that can be harvested. Solar absorbers are used in industrial environments to absorb the sun's radiation and turn it into thermal energy. This thermal energy can then be used to power things such as heating and cooling systems, air compressors, and even some types of manufacturing operations. By using a solar deep learning model, businesses can accurately predict how much thermal energy can be harvested from a particular solar absorber before making an investment in a system.

CURTAIN-A unique web-based tool for exploration and sharing of MS-based proteomics data.

To facilitate analysis and sharing of mass spectrometry (MS)-based proteomics data, we created online tools called CURTAIN (https://curtain.proteo.info) and CURTAIN-PTM (https://curtainptm.proteo.info) with an accompanying series of video tutorials (https://www.youtube.com/@CURTAIN-me6hl). These are designed to enable non-MS experts to interactively peruse volcano plots and deconvolute primary experimental data so that replicates can be visualized in bar charts or violin plots and exported in publication-ready format. They also allow assessment of overall experimental quality by correlation matrix and profile plot analysis. After making a selection of protein "hits", the user can analyze known domain structure, AlphaFold predicted structure, reported interactors, relative expression as well as disease links. CURTAIN-PTM permits analysis of all identified PTM sites on protein(s) of interest with selected databases. CURTAIN-PTM also links with the Kinase Library to predict upstream kinases that may phosphorylate sites of interest. We provide examples of the utility of CURTAIN and CURTAIN-PTM in analyzing how targeted degradation of the PPM1H Rab phosphatase that counteracts the Parkinson's LRRK2 kinase impacts cellular protein levels and phosphorylation sites. We also reanalyzed a ubiquitylation dataset, characterizing the PINK1-Parkin pathway activation in primary neurons, revealing data of interest not highlighted previously. CURTAIN and CURTAIN-PTM are free to use and open source, enabling researchers to share and maximize the impact of their proteomics data. We advocate that MS data published in volcano plot format be reported containing a shareable CURTAIN weblink, thereby allowing readers to better analyze and exploit the data.

PolyGR and polyPR knock-in mice reveal a conserved neuroprotective extracellular matrix signature in C9orf72 ALS/FTD neurons.

Dipeptide repeat proteins are a major pathogenic feature of C9orf72 amyotrophic lateral sclerosis (C9ALS)/frontotemporal dementia (FTD) pathology, but their physiological impact has yet to be fully determined. Here we generated C9orf72 dipeptide repeat knock-in mouse models characterized by expression of 400 codon-optimized polyGR or polyPR repeats, and heterozygous C9orf72 reduction. (GR)400 and (PR)400 knock-in mice recapitulate key features of C9ALS/FTD, including cortical neuronal hyperexcitability, age-dependent spinal motor neuron loss and progressive motor dysfunction. Quantitative proteomics revealed an increase in extracellular matrix (ECM) proteins in (GR)400 and (PR)400 spinal cord, with the collagen COL6A1 the most increased protein. TGF-ß1 was one of the top predicted regulators of this ECM signature and polyGR expression in human induced pluripotent stem cell neurons was sufficient to induce TGF-ß1 followed by COL6A1. Knockdown of TGF-ß1 or COL6A1 orthologues in polyGR model Drosophila exacerbated neurodegeneration, while expression of TGF-ß1 or COL6A1 in induced pluripotent stem cell-derived motor neurons of patients with C9ALS/FTD protected against glutamate-induced cell death. Altogether, our findings reveal a neuroprotective and conserved ECM signature in C9ALS/FTD.

Assessment of the three-test genetic toxicology battery for groundwater metabolites.

The two-test in vitro battery for genotoxicity testing (Ames and micronucleus) has in the majority of cases replaced the three-test battery (as two-test plus mammalian cell gene mutation assay) for the routine testing of chemicals, pharmaceuticals, cosmetics, and agrochemical metabolites originating from food and feed as well as from water treatment. The guidance for testing agrochemical groundwater metabolites, however, still relies on the three-test battery. Data collated in this study from 18 plant protection and related materials highlights the disparity between the often negative Ames and in vitro chromosome aberration data and frequently positive in vitro mammalian cell gene mutation assays. Sixteen of the 18 collated materials with complete datasets were Ames negative, and overall had negative outcomes in in vitro chromosome damage tests (weight of evidence from multiple tests). Mammalian cell gene mutation assays (HPRT and/or mouse lymphoma assay (MLA)) were positive in at least one test for every material with this data. Where both MLA and HPRT tests were performed on the same material, the HPRT seemed to give fewer positive responses. In vivo follow-up tests included combinations of comet assays, unscheduled DNA synthesis, and transgenic rodent gene mutation assays, all gave negative outcomes. The inclusion of mammalian cell gene mutation assays in a three-test battery for groundwater metabolites is therefore not justified and leads to unnecessary in vivo follow-up testing.

Parkinson's VPS35[D620N] mutation induces LRRK2-mediated lysosomal association of RILPL1 and TMEM55B.

We demonstrate that the Parkinson's VPS35[D620N] mutation alters the expression of ~220 lysosomal proteins and stimulates recruitment and phosphorylation of Rab proteins at the lysosome. This recruits the phospho-Rab effector protein RILPL1 to the lysosome where it binds to the lysosomal integral membrane protein TMEM55B. We identify highly conserved regions of RILPL1 and TMEM55B that interact and design mutations that block binding. In mouse fibroblasts, brain, and lung, we demonstrate that the VPS35[D620N] mutation reduces RILPL1 levels, in a manner reversed by LRRK2 inhibition and proteasome inhibitors. Knockout of RILPL1 enhances phosphorylation of Rab substrates, and knockout of TMEM55B increases RILPL1 levels. The lysosomotropic agent LLOMe also induced LRRK2 kinase-mediated association of RILPL1 to the lysosome, but to a lower extent than the D620N mutation. Our study uncovers a pathway through which dysfunctional lysosomes resulting from the VPS35[D620N] mutation recruit and activate LRRK2 on the lysosomal surface, driving assembly of the RILPL1-TMEM55B complex.

Early and Mid-Term Outcomes of Primary Repair After Atrioventricular Canal Defect: A Single-Center Eight-Year Experience.

BACKGROUND/AIM: Surgical repair techniques and management of patients with atrioventricular septal defect (AVSD) have progressed over the last few decades. Early and definitive interventions have become the choice of treatment for these patients. Based on this background, we aimed to review the early and mid-term outcomes of primary AVSD repair. METHODS: A total of 53 patients with a mean age of 3.45 ± 5.67 years underwent definitive repair for AVSD between January 2014 and June 2021. The clinical data including age, type of defect, associated co-anomalies, symptoms, pulmonary hypertension, etc. were collected and assessed retrospectively. Mitral regurgitation (MR) as a clinical outcome was assessed at 0, 1, 2, and 5 years. RESULTS: Among the recruited patients, 35 (66.1%) were male and 18 (33.9%) were female. Of 53 patients, repair for the complete defect was done in 38 (71.69%) patients, repair for intermediate/partial defect was done in 15 (23.1%) patients, and one patient underwent repair for incomplete type. Other associated co-anomalies were anterior mitral leaflet (12 (22.6%)), atrial and ventricular septal defect (VSD) (30 (56.6%)), and patent ductus arteriosus (PDA) (11 (20.8%)). Different procedures for surgical repair included patch closure, cleft repair, and polytetrafluoroethylene (PTFE) VSD closure. After repair, the mean follow-up period was 46.73 ± 27.37 months. Overall mortality was 3.78% (2/53), and two patients underwent reintervention due to symptomatic severe MR. CONCLUSIONS: A definitive and timely correction of AVSD shows satisfactory early and mid-term results.

Effectiveness and Safety of the Ozaki Procedure for Aortic Valve Disease in Pediatric Patients: A Systematic Review and Meta-Analysis.

The surgical treatment options for pediatric aortic valve disease are limited and have debatable long-term durability. In the current situation, the Ross procedure is considered in children for aortic valve disease(s). It is a complex surgical procedure with the risk of neo-aortic dilatation, converting a single valve disease into double valve disease, and associated with future re-interventions. Conversely, the Ozaki procedure has shown promising results in adults. Thus, the present study aimed to provide comparative evidence on the effectiveness and safety of the Ozaki versus Ross procedure for pediatric patients by performing a meta-analytic comparison of reporting outcomes. A total of 15 relevant articles were downloaded and among them, seven articles (one prospective study, five retrospective studies, and one case series) were used in the analysis. Primary outcomes such as physiological laminar flow pattern and hemodynamic parameters, and secondary outcomes such as hospital stays, adverse effects, mortality, and numbers of re-intervention(s) were measured in the meta-analysis. There were no significant differences in the age of patients between children who underwent the Ozaki procedure and those who underwent the Ross procedure at the time of surgeries. The Ozaki procedure is a good solution to an aortic problem(s) similar to the Ross procedure. Unlike the Ross procedure, the Ozaki procedure has restored a physiological laminar flow pattern in the short-term follow-up without the bi-valvular disease. Mean hospital stays (p = 0.048), mean follow-up (p = 0.02), adverse effects (p = 0.02), death, and numbers of re-intervention(s) of children who underwent the Ozaki procedure were fewer than those who underwent the Ross procedure. The time required for re-intervention(s) is higher for children who underwent the Ozaki procedure than those who underwent the Ross procedure. None of the procedures, including the Ozaki procedure for aortic valve disease(s), has significant effects on hemodynamic parameters and the frequent death rate of children after surgeries. Based on our analysis, we may suggest the Ozaki procedure for aortic valve disease surgery in children.

Pumping machine fault diagnosis based on fused RDC-RBF.

At present, the fault diagnosis of pumping units in major oil fields in China is time-consuming and inefficient, and there is no universal problem for high requirements of hardware resources. In this study, a fault fusion diagnosis method of pumping unit based on improved Fourier descriptor (IDF) and rapid density clustering RBF (RDC-RBF) neural network is proposed. Firstly, the minimum inertia axis of the center of gravity of the indicator diagram is obtained. The farthest point of the intersection of the inertial axis and the contour is determined as the starting point. Then Fourier transform is performed on the contour boundary of the graph to obtain the feature vector. Then, combining with the idea of fast density clustering algorithm, the number of hidden layer neurons of RBF is determined by finding the point with the highest density and using it as the hidden layer neuron. At the same time, the characteristics of Gaussian function are introduced to ensure the activity of hidden layer neurons. Finally, through dynamic adaptive cuckoo search (DACS), the step size is automatically adjusted according to the convergence speed of the objective function of RBF, and the efficiency and accuracy of RBF in different search stages are balanced. The optimal parameters such as the width and weight of RBF are determined, and the optimal RDC-RBF fault diagnosis model is established. The model is applied to the diagnosis of different fault types of pumping units, and compared with the current mainstream models. The average detection accuracy of the fusion RDC-RBF fault diagnosis method proposed in this paper reaches 96.3%. The measured results have high accuracy and short time. At the same time, this method is currently applied to oil production sites such as Shengli Oilfield in China, which greatly reduces the human resources required for fault diagnosis of pumping units in the past.

Endovascular Stenting for May-Thurner Syndrome: A Case Report.

May-Thurner syndrome, also known as iliocaval compression syndrome, is a rare vascular condition that involves compression of the left common iliac vein by the right common iliac artery. This compression can lead to venous stasis and increase the risk of deep vein thrombosis in the left lower extremity. Treatment options range from conservative measures to endovascular procedures such as venous stenting. Here, we present the case of a 45-year-old female with a history of recurrent deep vein thrombosis in her left leg, who arrived at the emergency department with swelling, pain, and tenderness. She was on warfarin therapy for deep vein thrombosis management. Physical examination and laboratory investigations supported the diagnosis of acute deep vein thrombosis. Further investigations revealed May-Thurner syndrome, with the left common iliac vein being compressed by the right common iliac artery, leading to extensive thrombosis in the left lower extremity. Endovascular stenting was performed to relieve the obstruction and restore venous blood flow. The patient's symptoms improved after the stenting procedure, and she remained asymptomatic during follow-up with continued anticoagulation therapy. Awareness of May-Thurner syndrome is crucial, especially in patients with recurrent deep venous thrombosis and anatomical risk factors. Successful management requires a multidisciplinary approach involving anticoagulation therapy and endovascular stenting.

Error-corrected next generation sequencing - Promises and challenges for genotoxicity and cancer risk assessment.

Error-corrected Next Generation Sequencing (ecNGS) is rapidly emerging as a valuable, highly sensitive and accurate method for detecting and characterizing mutations in any cell type, tissue or organism from which DNA can be isolated. Recent mutagenicity and carcinogenicity studies have used ecNGS to quantify drug-/chemical-induced mutations and mutational spectra associated with cancer risk. ecNGS has potential applications in genotoxicity assessment as a new readout for traditional models, for mutagenesis studies in 3D organotypic cultures, and for detecting off-target effects of gene editing tools. Additionally, early data suggest that ecNGS can measure clonal expansion of mutations as a mechanism-agnostic early marker of carcinogenic potential and can evaluate mutational load directly in human biomonitoring studies. In this review, we discuss promising applications, challenges, limitations, and key data initiatives needed to enable regulatory testing and adoption of ecNGS - including for advancing safety assessment, augmenting weight-of-evidence for mutagenicity and carcinogenicity mechanisms, identifying early biomarkers of cancer risk, and managing human health risk from chemical exposures.

Burkholderia cepacia infections at sites other than the respiratory tract: A large case series from a tertiary hospital in Madurai.

Sparsely reported extrapulmonary Burkholderia cepacia complex (Bcc) infections highlights the importance of this study. This was a retrospective chart review of 37 patients with extrapulmonary Bcc infections admitted between December 2019 and July 2022 in a tertiary hospital. Males accounted for 70% of cases. 78% had atleast one underlying comorbid illness. Among 37 isolates, 22 were from blood, others include exudates, urine and peritoneal fluid. Susceptibility rates of ceftazidime, meropenem, minocycline, cotrimoxazole and levofloxacin were 88, 88, 70, 65.7 and 56.7% respectively. Eleven died of septic shock and 24 patients (64.8%) had good outcomes, while two were lost to followup.

Telomere biology disorder presenting acutely with pulmonary fibrosis and hepatopulmonary syndrome in a young adult male.

A 33-year-old man presented with acute dyspnoea and profound hypoxaemia, and had clubbing, greying of hair, orthodeoxia and fine inspiratory crackles. CT chest showed established pulmonary fibrosis in a usual interstitial pneumonia pattern. Additional investigations revealed a small patent foramen ovale, pancytopenia, and oesophageal varices and portal hypertensive gastropathy from liver cirrhosis. Telomere length testing demonstrated short telomeres (<1st percentile), confirming the diagnosis of a telomere biology disorder. An interstitial lung disease gene panel identified a pathogenic variant in TERT (c.1700C>T, p.(Thr567Met)) and a variant of uncertain significance in PARN (c.1159G>A, p.(Gly387Arg)). Combined lung and liver transplantation was deemed not suitable due to frailty and severe hepatopulmonary syndrome, and he died 56 days after presentation. Early recognition of the short telomere syndrome is important, and its multi-organ involvement poses challenges to management. Genetic screening may be important in younger patients with pulmonary fibrosis or in unexplained liver cirrhosis.