Diagnostics

In recent times, DFU (diabetic foot ulcer) has become a universal health problem that affects many diabetes patients severely. DFU requires immediate proper treatment to avert amputation. Clinical examination of DFU is a tedious process and complex in nature. Concurrently, DL (deep learning) methodologies can show prominent outcomes in the classification of DFU because of their efficient learning capacity. Though traditional systems have tried using DL-based models to procure better performance, there is room for enhancement in accuracy. Therefore, the present study uses the AWSg-CNN (Adaptive Weighted Sub-gradient Convolutional Neural Network) method to classify DFU. A DFUC dataset is considered, and several processes are involved in the present study. Initially, the proposed method starts with pre-processing, excluding inconsistent and missing data, to enhance dataset quality and accuracy. Further, for classification, the proposed method utilizes the process of RIW (random initialization of weights) and log softmax with the ASGO (Adaptive Sub-gradient Optimizer) for effective performance. In this process, RIW efficiently learns the shift of feature space between the convolutional layers. To evade the underflow of gradients, the log softmax function is used. When logging softmax with the ASGO is used for the activation function, the gradient steps are controlled. An adaptive modification of the proximal function simplifies the learning rate significantly, and optimal proximal functions are produced. Due to such merits, the proposed method can perform better classification. The predicted results are displayed on the webpage through the HTML, CSS, and Flask frameworks. The effectiveness of the proposed system is evaluated with accuracy, recall, F1-score, and precision to confirm its effectual performance. Full article

Pediatric hemato-oncology patients undergoing anthracycline therapy are at risk of cardiotoxicity, with disease type and treatment intensity potentially affecting cardiac function. Novel echocardiographic measures like speckle tracking echocardiography (STE), global longitudinal strain (GLS), and the myocardial performance index (MPI) may predict early changes in cardiac function not detected by traditional methods. This study aimed to assess the impact of cancer type and treatment protocol on these parameters and their potential in predicting long-term cardiac complications. We conducted a single-center, retrospective cohort study of 99 pediatric oncology patients and 46 controls that were assessed at 3, 6, and 12 months. The median age was 10.7 ± 4.4 years for cases and 10.2 ± 3.6 years for controls. STE, GLS, and MPI were measured, and statistical analyses were performed to determine any significant correlations with cardiotoxicity. Significant variations were observed in traditional cardiac function measurements between the patient and control groups, with a lower average ejection fraction (EF) of 62.8 ± 5.7% in patients vs. 66.4 ± 6.1% in controls (p < 0.001), poorer GLS of −16.3 ± 5.1 in patients compared to −19.0 ± 5.4 in controls (p = 0.004), and higher MPI values of 0.37 ± 0.06 in patients compared to 0.55 ± 0.10 in controls, indicating worse overall cardiac function (p < 0.001). However, differences in cardiac function measurements by cancer histology or treatment protocol were not statistically significant. Regression analyses showed that the combination of GLS, SMOD, and MPI increased the odds of cardiac toxicity with an odds ratio of 7.30 (95% CI: 2.65–12.81, p < 0.001). The study underscores the predictive value of the combined GLS, SMOD, and MPI measurements in pediatric oncology patients undergoing anthracycline treatment for cardiotoxicity. Although variations across cancer types and treatment protocols were not significant, the study emphasizes the potential utility of these novel echocardiographic measures in early detection and long-term prediction of anthracycline-induced cardiotoxicity. Further studies in larger, multi-center cohorts are required for validation. Full article

Background: It has been reported that mid-regional proadrenomedullin (MR-proADM) could be considered a useful tool to stratify the mortality risk in COVID-19 patients upon admission to the emergency department (ED). During the COVID-19 outbreak, computed tomography (CT) scans were widely used for their excellent sensitivity in diagnosing pneumonia associated with SARS-CoV-2 infection. However, the possible role of CT score in the risk stratification of COVID-19 patients upon admission to the ED is still unclear. Aim: The main objective of this study was to assess if the association of the CT findings alone or together with MR-proADM results could ameliorate the prediction of in-hospital mortality of COVID-19 patients at the triage. Moreover, the hypothesis that CT score and MR-proADM levels together could play a key role in predicting the correct clinical setting for these patients was also evaluated. Methods: Epidemiological, demographic, clinical, laboratory, and outcome data were assessed and analyzed from 265 consecutive patients admitted to the triage of the ED with a SARS-CoV-2 infection. Results and conclusions: The accuracy results by AUROC analysis and statistical analysis demonstrated that CT score is particularly effective, when utilized together with the MR-proADM level, in the risk stratification of COVID-19 patients admitted to the ED, thus helping the decision-making process of emergency physicians and optimizing the hospital resources. Full article

The acquisition of ultrasound skills is an essential part of any medical student’s education. University access restrictions in the context of the COVID-19 pandemic have highlighted the need for digitization in teaching. However, teaching manual skills in online courses has proven to be challenging, not least in terms of human resources. Therefore, the aim of this study was to set up a hybrid head and neck ultrasound course consisting of a preface of video-based self-study followed by supportive instruction by a tutor in telepresence and to evaluate the quality, effectiveness, and feasibility of this teaching method. Thirty-five students were shown video tutorials on systematic ultrasound of the neck course. Learning outcomes were analyzed using self-assessment questionnaires and external assessment by an experienced ultrasonographer. All participants demonstrated statistically significant learning improvement (p < 0.001) when comparing self-assessment scores before and after training. The mean self-assessment scores increased from 13.8 to 26.6 for the telepresence-guided group, from 16.6 to 27.3 for the web-based group, and from 14.0 to 26.2 for the in-person group. The external observer assessment also showed improvement, with mean scores of 46.7, 48.1, and 46.5, respectively. Overall results did not significantly differ when comparing different instruction modalities. A telepresence-guided video-based ultrasound course is well suited to teaching ultrasound skills similar to in-person courses and allows a more resource-efficient targeting of student needs. Full article

Parkinson’s disease (PD) is a chronic and progressive neurological disease that mostly shakes and compromises the motor system of the human brain. Patients with PD can face resting tremors, loss of balance, bradykinesia, and rigidity problems. Complex patterns of PD, i.e., with relevance to other neurological diseases and minor changes in brain structure, make the diagnosis of this disease a challenge and cause inaccuracy of about 25% in the diagnostics. The research community utilizes different machine learning techniques for diagnosis using handcrafted features. This paper proposes a computer-aided diagnostic system using a convolutional neural network (CNN) to diagnose PD. CNN is one of the most suitable models to extract and learn the essential features of a problem. The dataset is obtained from Parkinson’s Progression Markers Initiative (PPMI), which provides different datasets (benchmarks), such as T2-weighted MRI for PD and other healthy controls (HC). The mid slices are collected from each MRI. Further, these slices are registered for alignment. Since the PD can be found in substantia nigra (i.e., the midbrain), the midbrain region of the registered T2-weighted MRI slice is selected using the freehand region of interest technique with a 33 × 33 sized window. Several experiments have been carried out to ensure the validity of the CNN. The standard measures, such as accuracy, sensitivity, specificity, and area under the curve, are used to evaluate the proposed system. The evaluation results show that CNN provides better accuracy than machine learning techniques, such as naive Bayes, decision tree, support vector machine, and artificial neural network. Full article

When hypostatic pneumonia is present at the same time as COVID-19 pneumonia, the clinical course is almost always prolonged (prolonged-COVID-19) due to persistent inflammation, long-term anti-inflammatory syndrome, followed by immune exhaustion, i.e., by immunosuppression and catabolic syndrome. In the immunosuppression phase, viral reactivation can be accompanied by a secondary infection, which, in this case, is pulmonary tuberculosis. Pulmonary tuberculosis in post-COVID-19 patients and in patients with spastic quadriplegic cerebral palsy does not have a typical clinical course nor laboratory, radiological, immunological, microbiological, or fiberbronchoscopic pathohistological confirmation. Due to this, the treatment of pulmonary tuberculosis was not carried out on time, postponed after the unsuccessful treatment of sepsis, post-COVID-19, and other accompanying viral (adenovirus, RSV) and bacterial (streptococcus viridans) infections. The treatment of pulmonary tuberculosis was possible only “ex juvantibus” (trial) post-COVID-19. It becomes imperative to search for a new, more precise and reliable diagnostic test for the detection of tuberculosis bacillus. Full article

In hospitals and other clinical settings, Methicillin-resistant Staphylococcus aureus (MRSA) is a particularly dangerous pathogen that can cause serious or even fatal infections. Thus, the detection and differentiation of MRSA has become an urgent matter in order to provide appropriate treatment and timely intervention in infection control. To ensure this, laboratories must have access to the most up-to-date testing methods and technology available. This study was conducted to determine whether protein fingerprinting technology could be used to identify and distinguish MRSA recovered from both inpatients and outpatients. A total of 326 S. aureus isolates were obtained from 2800 in- and outpatient samples collected from King Faisal Specialist Hospital and Research Centre in Riyadh, Saudi Arabia, from October 2018 to March 2021. For the phenotypic identification of 326 probable S. aureus cultures, microscopic analysis, Gram staining, a tube coagulase test, a Staph ID 32 API system, and a Vitek 2 Compact system were used. Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS), referred to as protein fingerprinting, was performed on each bacterial isolate to determine its proteomic composition. As part of the analysis, Principal Component Analysis (PCA) and a single-peak analysis of MALDI-TOF MS software were also used to distinguish between Methicillin-sensitive Staphylococcus aureus (MSSA) and MRSA. According to the results, S. aureus isolates constituted 326 out of 2800 (11.64%) based on the culture technique. The Staph ID 32 API system and Vitek 2 Compact System were able to correctly identify 262 (80.7%) and 281 (86.2%) S. aureus strains, respectively. Based on the Oxacillin Disc Diffusion Method, 197 (62.23%) of 326 isolates of S. aureus exhibited a cefoxitin inhibition zone of less than 21 mm and an oxacillin inhibition zone of less than 10 mm, and were classified as MRSA under Clinical Laboratory Standards Institute guidelines. MALDI-TOF MS was able to correctly identify 100% of all S. aureus isolates with a score value equal to or greater than 2.00. In addition, a close relationship was found between S. aureus isolates and higher peak intensities in the mass ranges of 3990 Da, 4120 Da, and 5850 Da, which were found in MRSA isolates but absent in MSSA isolates. Therefore, protein fingerprinting has the potential to be used in clinical settings to rapidly detect and differentiate MRSA isolates, allowing for more targeted treatments and improved patient outcomes. Full article

The aim of this clinical study was to compare the diagnostic performance of dual short wavelength infrared (SWIR) occlusal transillumination and reflectance multispectral imaging with conventional visual assessment and radiography for caries detection on premolars scheduled for extraction for orthodontics reasons. Polarized light microscopy (PLM) and micro-computed tomography (microCT) performed after tooth extraction were used as gold standards. The custom-fabricated imaging probe was 3D-printed and the imaging system employed a SWIR camera and fiber-optic light sources emitting light at 1300 nm for occlusal transillumination and 1600 nm for reflectance measurements. Teeth (n = 135) on 40 test subjects were imaged in vivo using the SWIR imaging prototype in the study and teeth were extracted after imaging. Our study demonstrates for the first time that near-simultaneous real-time transillumination and reflectance video can be successfully acquired for caries detection. Both SWIR imaging modalities had markedly higher sensitivity for lesions on proximal and occlusal surfaces compared to conventional methods (visual and radiographic). Reflectance imaging at 1600 nm had higher sensitivity and specificity than transillumination at 1300 nm. The combined SWIR methods yielded higher specificity but the combined sensitivity was lower than for each individual method. Full article

The complete diagnostic evaluation of tuberculosis based on its drug-resistance profile is critical for appropriate treatment decisions. The TB diagnostic landscape in India has been transformed with the scaling-up of WHO-recommended diagnostics, but challenges remain with specimen transportation, completing diagnostic assessment, turnaround time (TAT), and maintaining laboratories. Private laboratories have demonstrated efficiencies for specimen collection, transportation, and the timely testing and issue of results. A one-stop TB diagnostic model was designed to assess the feasibility of providing end-to-end diagnostic services in the Hisar district of Haryana state, India. A NTEP-certified private laboratory was engaged to provide the services, complementing the existing public sector diagnostic services. A total of 10,164 specimens were collected between May 2022 and January 2023 and these were followed for the complete diagnostic assessment of Drug-Susceptible TB (DS-TB) and Drug-Resistant TB (DR-TB) and the time taken for issuing results. A total of 2152 (21%) patients were detected with TB, 1996 (93%) Rifampicin-Sensitive and 134 (6%) with Rifampicin-Resistant TB. Nearly 99% of the patients completed the evaluation of DS-TB and DR-TB within the recommended TAT. The One-Stop TB/DR-TB Diagnostic Solution model has demonstrated that diagnostic efficiencies could be enhanced through the strategic purchase of private laboratory services. Full article

Lymphedema is a pathology caused by poor lymphatic flow which may lead to complete disability. Currently, precise, non-invasive techniques for quantifying lymphedema are lacking. In this paper, the results of an in vivo assessment of lymphedema via a developed small-animal model using the hindlimbs of rats and an optical coherence tomography (OCT) technique are presented. This model of lymphedema was based on a surgical lymph node resection and subsequent two-step X-ray exposure. The development of lymphedema was verified via the histological examination of tissue biopsies. The properties of the lymphedematous skin were analyzed in vivo and compared with healthy skin via OCT. The main differences observed were (1) a thickening of the stratum corneum layer, (2) a thinning of the viable epidermis layer, and (3) higher signal attenuation in the dermis layer of the lymphedematous skin. Based on the distribution of the OCT signal’s intensity in the skin, a machine learning algorithm was developed which allowed for a classification of normal and lymphedematous tissue sites with an accuracy of 90%. The obtained results pave the way for in vivo control over the development of lymphedema. Full article

This study aimed to compare the image quality and diagnostic accuracy of deep-learning-based image denoising reconstructions (DLIDs) to established iterative reconstructed algorithms in low-dose computed tomography (LDCT) of patients with suspected urolithiasis. LDCTs (CTDIvol, 2 mGy) of 76 patients (age: 40.3 ± 5.2 years, M/W: 51/25) with suspected urolithiasis were retrospectively included. Filtered-back projection (FBP), hybrid iterative and model-based iterative reconstruction (HIR/MBIR, respectively) were reconstructed. FBP images were processed using a Food and Drug Administration (FDA)-approved DLID. ROIs were placed in renal parenchyma, fat, muscle and urinary bladder. Signal- and contrast-to-noise ratios (SNR/CNR, respectively) were calculated. Two radiologists evaluated image quality on five-point Likert scales and urinary stones. The results showed a progressive decrease in image noise from FBP, HIR and DLID to MBIR with significant differences between each method (p < 0.05). SNR and CNR were comparable between MBIR and DLID, while it was significantly lower in HIR followed by FBP (e.g., SNR: 1.5 ± 0.3; 1.4 ± 0.4; 1.0 ± 0.3; 0.7 ± 0.2, p < 0.05). Subjective analysis confirmed best image quality in MBIR, followed by DLID and HIR, both being superior to FBP (p < 0.05). Diagnostic accuracy for urinary stone detection was best using MBIR (0.94), lowest using FBP (0.84) and comparable between DLID (0.90) and HIR (0.90). Stone size measurements were consistent between all reconstructions and showed excellent correlation (r² = 0.958–0.975). In conclusion, MBIR yielded the highest image quality and diagnostic accuracy, with DLID producing better results than HIR and FBP in image quality and matching HIR in diagnostic precision. Full article

Neuroendocrine neoplasms (NENs) are a heterogeneous group of diseases that are characterized by different behavior and clinical manifestations. The diagnosis and management of this group of tumors are challenging due to tumor complexity and lack of precise and widely validated biomarkers. Indeed, the current circulating mono-analyte biomarkers (such as chromogranin A) are ineffective in describing such complex tumors due to their poor sensitivity and specificity. In contrast, multi-analytical circulating biomarkers (including NETest) are emerging as more effective tools to determine the real-time profile of the disease, both in terms of accurate diagnosis and effective treatment. In this review, we will analyze the capabilities and limitations of different circulating biomarkers focusing on three relevant questions: (1) accurate and early diagnosis; (2) monitoring of disease progression and response to therapy; and (3) detection of early relapse. Full article

The use of ultrasonography to predict spinal-induced hypotension (SIH) has gained significant attention. This diagnostic meta-analysis aimed to investigate the reliability of the inferior vena cava collapsibility index (IVCCI) in predicting SIH in patients undergoing various surgeries. Databases, including Embase, Cochrane Library, Medline, and Google Scholar, were screened until 28 July 2023, yielding 12 studies with 1076 patients (age range: 25.6–79 years) undergoing cesarean section (CS) (n = 4) or non-CS surgeries (n = 8). Patients with SIH had a significantly higher IVCCI than those without SIH (mean difference: 11.12%, 95% confidence interval (CI): 7.83–14.41). The pooled incidence rate of SIH was 40.5%. IVCCI demonstrated satisfactory overall diagnostic reliability (sensitivity, 77%; specificity, 82%). The pooled area under the curve (AUC) was 0.85, indicating its high capability to differentiate patients at risk of PSH. The Fagan nomogram plot demonstrated a positive likelihood ratio (PLR) of 4 and a negative likelihood ratio (NLR) of 0.28. The results underscore the robustness and discriminative ability of IVCCI as a predictive tool for SIH. Nevertheless, future investigations should focus on assessing its applicability to high-risk patients and exploring the potential enhancement in patient safety through its incorporation into clinical practice. Full article

Heart transplantation (HT) is the established treatment for end-stage heart failure, significantly enhancing patients’ survival and quality of life. To ensure optimal outcomes, the routine monitoring of HT recipients is paramount. While existing guidelines offer guidance on a blend of invasive and non-invasive imaging techniques, certain aspects such as the timing of echocardiographic assessments and the role of echocardiography or cardiac magnetic resonance (CMR) as alternatives to serial endomyocardial biopsies (EMBs) for rejection monitoring are not specifically outlined in the guidelines. Furthermore, invasive coronary angiography (ICA) is still recommended as the gold-standard procedure, usually performed one year after surgery and every two years thereafter. This review focuses on recent advancements in non-invasive and contrast-saving imaging techniques that have been investigated for HT patients. The aim of the manuscript is to identify imaging modalities that may potentially replace or reduce the need for invasive procedures such as ICA and EMB, considering their respective advantages and disadvantages. We emphasize the transformative potential of non-invasive techniques in elevating patient care. Advanced echocardiography techniques, including strain imaging and tissue Doppler imaging, offer enhanced insights into cardiac function, while CMR, through its multi-parametric mapping techniques, such as T1 and T2 mapping, allows for the non-invasive assessment of inflammation and tissue characterization. Cardiac computed tomography (CCT), particularly with its ability to evaluate coronary artery disease and assess graft vasculopathy, emerges as an integral tool in the follow-up of HT patients. Recent studies have highlighted the potential of nuclear myocardial perfusion imaging, including myocardial blood flow quantification, as a non-invasive method for diagnosing and prognosticating CAV. These advanced imaging approaches hold promise in mitigating the need for invasive procedures like ICA and EMB when evaluating the benefits and limitations of each modality. Full article

Left ventricular diastolic dysfunction (DD) is a subclinical cardiac abnormality in patients with type 2 diabetes mellitus (T2DM) that can progress to heart failure (HF) and increase cardiovascular risk. This prospective study evaluated the DD in T2DM patients without atherosclerotic cardiovascular disease after one year of incretin-based drugs added to standard treatment. Of the 138 enrolled patients (49.30% male, mean age 57.86 ± 8.82, mean T2DM history 5 years), 71 were started on dipeptidyl peptidase-4 inhibitor sitagliptin/saxagliptin, 21 on glucagon-like peptide-1 receptor agonist exenatide, and 46 formed the control group (metformin and sulphonylurea/acarbose). At baseline, 71 patients had grade 1 DD, another 12 had grade 2 and 3 DD, and 15 had indeterminate DD. After one year, DD was evidenced in 50 cases. Diastolic function improved in 9 cases, and 27 patients went from grade 1 to indeterminate DD. The active group benefited more, especially patients treated with exenatide; their metabolic and inflammation profiles also improved the most. An in-depth analysis of echocardiographic parameters and paraclinical results in the context of literature data justifies the conclusion that early assessment of diastolic function in T2DM patients is necessary and the benefits of affordable incretin-based treatment may extend to subclinical cardiovascular manifestations such as DD. Full article

Parkinson’s disease (PD) is a neurodegenerative disorder marked by motor and non-motor symptoms that have a severe impact on the quality of life of the affected individuals. This study explores the effect of filter feature selection, followed by ensemble learning methods and genetic selection, on the detection of PD patients from attributes extracted from voice clips from both PD patients and healthy patients. Two distinct datasets were employed in this study. Filter feature selection was carried out by eliminating quasi-constant features. Several classification models were then tested on the filtered data. Decision tree, random forest, and XGBoost classifiers produced remarkable results, especially on Dataset 1, where 100% accuracy was achieved by decision tree and random forest. Ensemble learning methods (voting, stacking, and bagging) were then applied to the best-performing models to see whether the results could be enhanced further. Additionally, genetic selection was applied to the filtered data and evaluated using several classification models for their accuracy and precision. It was found that in most cases, the predictions for PD patients showed more precision than those for healthy individuals. The overall performance was also better on Dataset 1 than on Dataset 2, which had a greater number of features. Full article

Endoscopic ultrasound (EUS) has emerged as a widely utilized tool in the diagnosis of digestive diseases. In recent years, the potential of artificial intelligence (AI) in healthcare has been gradually recognized, and its superiority in the field of EUS is becoming apparent. Machine learning (ML) and deep learning (DL) are the two main AI algorithms. This paper aims to outline the applications and prospects of artificial intelligence-assisted endoscopic ultrasound (EUS-AI) in digestive diseases over the past decade. The results demonstrated that EUS-AI has shown superiority or at least equivalence to traditional methods in the diagnosis, prognosis, and quality control of subepithelial lesions, early esophageal cancer, early gastric cancer, and pancreatic diseases including pancreatic cystic lesions, autoimmune pancreatitis, and pancreatic cancer. The implementation of EUS-AI has opened up new avenues for individualized precision medicine and has introduced novel diagnostic and treatment approaches for digestive diseases. Full article

Background: We aimed to investigate the association between kidney stones and polycystic ovarian syndrome (PCOS). Materials and methods: In a cross-sectional study, data from the Tehran Lipid and Glucose Study (TLGS) were used to investigate the risk of kidney stones in women with Polycystic Ovary Syndrome (PCOS). Four distinct phenotypes of PCOS, as defined by the Rotterdam criteria, were examined in a sample of 520 women and compared to a control group of 1638 eumenorrheic non-hirsute healthy women. Univariate and multivariable logistic regression models were employed for analysis. The four PCOS phenotypes were classified as follows: Phenotype A, characterized by the presence of all three PCOS features (anovulation (OA), hyperandrogenism (HA), and polycystic ovarian morphology on ultrasound (PCOM)); Phenotype B, characterized by the presence of anovulation and hyperandrogenism; Phenotype C, characterized by the presence of hyperandrogenism and polycystic ovarian morphology on ultrasound; and Phenotype D, characterized by the presence of anovulation and polycystic ovarian morphology on ultrasound. Results: The prevalence of a history of kidney stones was found to be significantly higher in women with Polycystic Ovary Syndrome (PCOS) compared to healthy controls (12.5% vs. 7.7%, p = 0.001). This increased prevalence was observed across all PCOS phenotypes (p < 0.001). After adjusting for potential risk factors, including age, family history of kidney stones, waist-to-height ratio, total cholesterol, and low-density lipoprotein, the odds ratio for kidney stones in women with PCOS was found to be 1.59 [95% CI: 1.12–2.25, p = 0.01], indicating a 59% increase in risk compared to healthy women. Women with PCOS Phenotype A [OR: 1.97, 95% CI: 1.09–3.55, p = 0.02] and Phenotype D [OR: 3.03, 95% CI: 1.24–7.41, p = 0.01] were found to be at a higher risk for kidney stones. Conclusion: Women with Polycystic Ovary Syndrome (PCOS), particularly those exhibiting menstrual irregularities and polycystic ovarian morphology on ultrasound (PCOM), have been found to be two to three times more likely to develop kidney stones. This increased prevalence should be taken into consideration when providing preventive care and counseling to these individuals. Full article

Artificial intelligence (AI) methods applied to healthcare problems have shown enormous potential to alleviate the burden of health services worldwide and to improve the accuracy and reproducibility of predictions. In particular, developments in computer vision are creating a paradigm shift in the analysis of radiological images, where AI tools are already capable of automatically detecting and precisely delineating tumours. However, such tools are generally developed in technical departments that continue to be siloed from where the real benefit would be achieved with their usage. Significant effort still needs to be made to make these advancements available, first in academic clinical research and ultimately in the clinical setting. In this paper, we demonstrate a prototype pipeline based entirely on open-source software and free of cost to bridge this gap, simplifying the integration of tools and models developed within the AI community into the clinical research setting, ensuring an accessible platform with visualisation applications that allow end-users such as radiologists to view and interact with the outcome of these AI tools. Full article