MATERIALS AND METHODS: This was a retrospective study using computed tomography (CT) scans from 3 hospitals. Inclusion criteria were scans with 1-5 nodules of diameter ≥5 mm; exclusion criteria were poor-quality scans or those with nodules measuring <5mm in diameter. In the lesion detection phase, 2,147 nodules from 219 scans were used to develop and train the deep learning 3D-CNN to detect lesions. The 3D-CNN was validated with 235 scans (354 lesions) for sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) analysis. In the path planning phase, Bayesian optimization was used to propose possible needle trajectories for lesion biopsy while avoiding vital structures. Software-proposed needle trajectories were compared with actual biopsy path trajectories from intraprocedural CT scans in 150 patients, with a match defined as an angular deviation of <5° between the 2 trajectories.
RESULTS: The model achieved an overall AUC of 97.4% (95% CI, 96.3%-98.2%) for lesion detection, with mean sensitivity of 93.5% and mean specificity of 93.2%. Among the software-proposed needle trajectories, 85.3% were feasible, with 82% matching actual paths and similar performance between supine and prone/oblique patient orientations (P = .311). The mean angular deviation between matching trajectories was 2.30° (SD ± 1.22); the mean path deviation was 2.94 mm (SD ± 1.60).
CONCLUSIONS: Segmentation, lesion detection, and path planning for CT-guided lung biopsy using an AI-guided software showed promising results. Future integration with automated robotic systems may pave the way toward fully automated biopsy procedures.
Materials and Methods: All the variants' information was retrieved from the Ensembl genome database, and then different variation prediction analyses were performed. UTRScan was used to predict UTR variants while MaxEntScan was used to predict splice site variants. Meta-analysis by PredictSNP2 was used to predict sSNPs. Parallel prediction analyses by five different software packages including SIFT, PolyPhen-2, Mutation Assessor, I-Mutant2.0 and SNPs&GO were used to predict the effects of nsSNPs. The level of evolutionary conservation of deleterious nsSNPs was further analyzed using ConSurf server. Mutant protein structures of deleterious nsSNPs were modelled and refined using SPARKS-X and ModRefiner for structural comparison.
Results: A total of 56 deleterious variants were identified in this study, including 12 UTR variants, 18 splice site variants, eight sSNPs and 18 nsSNPs. Among these 56 deleterious variants, seven variants were also identified in the Alzheimer's Disease Sequencing Project (ADSP), Alzheimer's Disease Neuroimaging Initiative (ADNI) and Mount Sinai Brain Bank (MSBB) studies.
Discussion: The 56 deleterious variants were predicted to affect the regulation of gene expression, or have functional impacts on these three endocytosis genes and their gene products. The deleterious variants in these genes are expected to affect their cellular function in endocytosis and may be implicated in the pathogenesis of AD as well. The biological consequences of these deleterious variants and their potential impacts on the disease risks could be further validated experimentally and may be useful for gene-disease association study.
RESULT: An automated 3D modeling pipeline empowered by an Artificial Neural Network (ANN) was developed. This automated 3D modelling pipeline enables automated deformation of a generic 3D model of monogenean anchor into another target 3D anchor. The 3D modelling pipeline empowered by ANN has managed to automate the generation of the 8 target 3D models (representing 8 species: Dactylogyrus primaries, Pellucidhaptor merus, Dactylogyrus falcatus, Dactylogyrus vastator, Dactylogyrus pterocleidus, Dactylogyrus falciunguis, Chauhanellus auriculatum and Chauhanellus caelatus) of monogenean anchor from the respective 2D illustrations input without repeating the tedious modelling procedure.
CONCLUSIONS: Despite some constraints and limitation, the automated 3D modelling pipeline developed in this study has demonstrated a working idea of application of machine learning approach in a 3D modelling work. This study has not only developed an automated 3D modelling pipeline but also has demonstrated a cross-disciplinary research design that integrates machine learning into a specific domain of study such as 3D modelling of the biological structures.
METHODOLOGY: A cross-sectional survey using selfadministered questionnaires was conducted among all specialists working in government specialist hospitals in the northern states of Malaysia.
RESULTS: Out of 733 questionnaires distributed, 467 were returned giving a response rate of 63.7%. Ninety-nine percent of the respondents believed that research benefits patients while 93.3% think research helps in their professional development. However, 34.8% think that under their present working conditions, it is unlikely they will participate in research. The major barriers identified were lack of funds for research (81%); lack access to expertise, software or statistical analysis (78.4%); interference with daily work schedule (75.1%) and inconsistent manpower in their department (74.2%). There are three barriers with statistically significant difference between hospitals with CRC compared to hospitals without CRC; lack of funds, mentors and access to expertise, software or statistical analysis. The demographic factors, attitudes and barriers contributing to involvement in research also investigated. The main facilitators for the conduct of research are potential to benefit patients and potential for professional development.
CONCLUSION: Taking note of the findings, the Ministry of Health can implement appropriate strategies to improve specialist participation in research.
Aim: To determine the feasibility of a collaborative program between private general practitioners (GPs) and the public primary health clinics in PTB screening and to assess the yield of smear-positive PTB from this program.
Methods: A prospective cohort study using convenient sampling was conducted involving GPs and public health clinics in the North-East District, Penang, from March 2018 to May 2019. In this study, GPs could direct all suspected PTB patients to perform a sputum acid fast bacilli (AFB) direct smear in any of the dedicated public primary health clinics. The satisfaction level of both the GPs and their patients were assessed using a self-administered client satisfaction questionnaire. IBM SPSS Statistical Software was used to analyze the data.
Results: Out of a total of 31 patients who underwent the sputum investigation for PTB, one (3.2%) was diagnosed to have smear-positive PTB. Most of the patients (>90%) and GPs (66.7%) agreed to continue with this program in the future. Furthermore, most of the patients (>90%) were satisfied with the program structure.
Conclusion: It is potentially feasible to involve GPs in combating TB. However, a more structured program addressing the identified issues is needed to make the collaborative program a success.