Non-invasive diagnosis of deep vein thrombosis from ultrasound imaging with machine learning

Adopting artificial intelligence in cardiovascular medicine: a scoping review

Recent years have witnessed significant transformations in cardiovascular medicine, driven by the rapid evolution of artificial intelligence (AI). This scoping review was conducted to capture the breadth of AI applications within cardiovascular science. Employing a structured approach, we sourced relevant articles from PubMed, with an emphasis on journals encompassing general cardiology and digital medicine. We applied filters to highlight cardiovascular articles published in journals focusing on general internal medicine, cardiology and digital medicine, thereby identifying the prevailing trends in the field. Following a comprehensive full-text screening, a total of 140 studies were identified. Over the preceding 5 years, cardiovascular medicine's interplay with AI has seen an over tenfold augmentation. This expansive growth encompasses multiple cardiovascular subspecialties, including but not limited to, general cardiology, ischemic heart disease, heart failure, and arrhythmia. Deep learning emerged as the predominant methodology. The majority of AI endeavors in this domain have been channeled toward enhancing diagnostic and prognostic capabilities, utilizing resources such as hospital datasets, electrocardiograms, and echocardiography. A significant uptrend was observed in AI's application for omics data analysis. However, a clear gap persists in AI's full-scale integration into the clinical decision-making framework. AI, particularly deep learning, has demonstrated robust applications across cardiovascular subspecialties, indicating its transformative potential in this field. As we continue on this trajectory, ensuring the alignment of technological progress with medical ethics becomes crucial. The abundant digital health data today further accentuates the need for meticulous systematic reviews, tailoring them to each cardiovascular subspecialty.

Remote Expert DVT Triaging of Novice-User Compression Sonography with AI-Guidance

BACKGROUND

Compression ultrasonography of the leg is established for triaging proximal lower extremity deep vein thrombosis (DVT). AutoDVT, a machine-learning software, provides a tool for nonspecialists in acquiring compression sequences to be reviewed by an expert for patient triage. The purpose of this study was to test image acquisition and remote triaging in a clinical setting.

METHODS

Patients with a suspected DVT were recruited at 2 centers in Germany and Greece. Enrolled patients underwent an artificial intelligence-guided two-point compression examination by a nonspecialist using a handheld ultrasound device prior to a standard scan. Images collected by the software were uploaded for blind review by 5 qualified physicians. All reviewers rated the quality of all sequences on the American College of Emergency Physicians (ACEP) image quality scale (score 1-5, ≥ 3 defined as adequate imaging quality) and for an ACEP score ≥3, chose "Compressible", "Incompressible", or "Other". Sensitivity and specificity were calculated for adequate quality scans with an assessment as "Compressible" or "Incompressible". We define this group as diagnostic quality. To simulate a triaging clinical algorithm, a post hoc analysis was performed merging the "incomplete", the "low quality", and the "Incompressible" into a high-risk group for proximal DVT.

RESULTS

Seventy-three patients (average age 64.2 years, 44% females) were eligible for inclusion and scanned by 3 nonultrasound-qualified healthcare professionals. Three patients were excluded from further analysis due to incomplete scans. Sixty two of 70 (88.57%) of the completed scans were judged to be of adequate image quality with an average ACEP score of 3.35. Forty seven of 62 adequate AutoDVT scans were assessed as diagnostic quality, of which 8 were interpreted as positive for proximal DVT by the reviewers resulting in a sensitivity of 100% and specificity of 95.12%. When simulating a triaging algorithm, 34/73 (46.58%) of patients would be triaged as high risk and 8 would be confirmed as positive for proximal DVT (6 in the diagnostic and 2 in the low-quality cohort). Of 39/73 patients triaged as low risk, all were negative for proximal DVT in standard duplex; thus, this triaging algorithm could potentially save 53.42% of standard duplex scans.

CONCLUSIONS

Machine learning software was able to aid nonexperts in acquiring valid ultrasound images of venous compressions and allowed remote triaging. This strategy allows faster diagnosis and treatment of high-risk patients and can spare the need for multiple unnecessary duplex scans, the vast majority being negative.

A Quantitative Method for the Evaluation of Deep Vein Thrombosis in a Murine Model Using Three-Dimensional Ultrasound Imaging

Deep vein thrombosis (DVT) is a life-threatening condition that can lead to its sequelae pulmonary embolism (PE) or post-thrombotic syndrome (PTS). Murine models of DVT are frequently used in early-stage disease research and to assess potential therapies. This creates the need for the reliable and easy quantification of blood clots. In this paper, we present a novel high-frequency 3D ultrasound approach for the quantitative evaluation of the volume of DVT in an in vitro model and an in vivo murine model. The proposed method involves the use of a high-resolution ultrasound acquisition system and semiautomatic segmentation of the clot. The measured 3D volume of blood clots was validated to be correlated with in vitro blood clot weights with an R2 of 0.89. Additionally, the method was confirmed with an R2 of 0.91 in the in vivo mouse model with a cylindrical volume from macroscopic measurement. We anticipate that the proposed method will be useful in pharmacological or therapeutic studies in murine models of DVT.

Study on the safety and effectiveness of low-dose vs. regular-dose fondaparinux in preventing venous thromboembolism prophylaxis following total knee arthroplasty

Background

This study aims to evaluate the effectiveness and safety of low-dose (1.5 mg) fondaparinux for venous thromboembolism (VTE) prophylaxis in patients post-total knee arthroplasty (TKA).

Methods

We retrospectively identified 314 patients who carried out the primary TKAs and received fondaparinux for VTE chemoprophylaxis between July 2020 and December 2021. A total of 141 TKA patients were excluded according to the exclusion criteria. Two groups of patients were established: the low-dose group included 84 patients who injected 1.5 mg of fondaparinux, and the regular-dose group included 89 patients who injected 2.5 mg of fondaparinux. The pre-operative blood analysis and coagulation assays were performed. The surgical time, the incidence of symptomatic VET, blood loss, wound complication, bleeding, drainage, and mortality of patients were determined and assessed.

Results

The pre-operative blood analysis, body mass index, sex, age, and coagulation assays of patients in both groups were comparable. In terms of symptomatic pulmonary embolism and deep vein thrombosis, there was no significant difference (variation) between the two groups. However, patients in both groups showed a substantial difference in terms of blood loss, drain volume, wound complication, and transfusion rate.

Conclusion

In prevention of VET in patients post-TKA, low-dose fondaparin is as effective as conventional dose fondaparinux. A significant decrease in blood loss, post-surgical transfusion rates, and wound complications were detected in patients given low-dose fondaparinux compared to those receiving regular-dose fondaparinux.

Robust and Realtime Large Deformation Ultrasound Registration Using End-to-End Differentiable Displacement Optimisation

Image registration for temporal ultrasound sequences can be very beneficial for image-guided diagnostics and interventions. Cooperative human-machine systems that enable seamless assistance for both inexperienced and expert users during ultrasound examinations rely on robust, realtime motion estimation. Yet rapid and irregular motion patterns, varying image contrast and domain shifts in imaging devices pose a severe challenge to conventional realtime registration approaches. While learning-based registration networks have the promise of abstracting relevant features and delivering very fast inference times, they come at the potential risk of limited generalisation and robustness for unseen data; in particular, when trained with limited supervision. In this work, we demonstrate that these issues can be overcome by using end-to-end differentiable displacement optimisation. Our method involves a trainable feature backbone, a correlation layer that evaluates a large range of displacement options simultaneously and a differentiable regularisation module that ensures smooth and plausible deformation. In extensive experiments on public and private ultrasound datasets with very sparse ground truth annotation the method showed better generalisation abilities and overall accuracy than a VoxelMorph network with the same feature backbone, while being two times faster at inference.

In Situ Pulmonary Arterial Thrombosis-Literature Review and Clinical Significance of a Distinct Entity

Filling defects identified in the pulmonary arterial tree are commonly presumed to represent an embolic phenomenon originating from thrombi formed in remote veins, particularly lower-extremity deep venous thrombosis (DVT). However, accumulating evidence supports an underappreciated cause for pulmonary arterial thrombosis (PAT), namely, de novo thrombogenesis-where thrombosis arises within the pulmonary arteries in the absence of DVT. Although historically underrecognized, in situ PAT has become of heightened importance with the emergence of SARS-CoV-2 infection. In situ PAT is attributed to endothelial dysfunction, systemic inflammation, and acute lung injury, and has been described in a range of conditions including COVID-19, trauma, acute chest syndrome in sickle cell disease, pulmonary infections, and severe pulmonary arterial hypertension. The distinction between pulmonary embolus and in situ PAT may have important implications regarding management decisions and clinical outcomes. In this review, we summarize the pathophysiology, imaging appearances, and management of in situ PAT in various clinical situations. This understanding will promote optimal tailored treatment strategies for this increasingly recognized entity.

Assistive artificial intelligence for ultrasound image interpretation in regional anaesthesia: an external validation study

BACKGROUND

Ultrasonound is used to identify anatomical structures during regional anaesthesia and to guide needle insertion and injection of local anaesthetic. ScanNav Anatomy Peripheral Nerve Block (Intelligent Ultrasound, Cardiff, UK) is an artificial intelligence-based device that produces a colour overlay on real-time B-mode ultrasound to highlight anatomical structures of interest. We evaluated the accuracy of the artificial-intelligence colour overlay and its perceived influence on risk of adverse events or block failure.

METHODS

Ultrasound-guided regional anaesthesia experts acquired 720 videos from 40 volunteers (across nine anatomical regions) without using the device. The artificial-intelligence colour overlay was subsequently applied. Three more experts independently reviewed each video (with the original unmodified video) to assess accuracy of the colour overlay in relation to key anatomical structures (true positive/negative and false positive/negative) and the potential for highlighting to modify perceived risk of adverse events (needle trauma to nerves, arteries, pleura, and peritoneum) or block failure.

RESULTS

The artificial-intelligence models identified the structure of interest in 93.5% of cases (1519/1624), with a false-negative rate of 3.0% (48/1624) and a false-positive rate of 3.5% (57/1624). Highlighting was judged to reduce the risk of unwanted needle trauma to nerves, arteries, pleura, and peritoneum in 62.9-86.4% of cases (302/480 to 345/400), and to increase the risk in 0.0-1.7% (0/160 to 8/480). Risk of block failure was reported to be reduced in 81.3% of scans (585/720) and to be increased in 1.8% (13/720).

CONCLUSIONS

Artificial intelligence-based devices can potentially aid image acquisition and interpretation in ultrasound-guided regional anaesthesia. Further studies are necessary to demonstrate their effectiveness in supporting training and clinical practice.

CLINICAL TRIAL REGISTRATION

NCT04906018.

Evaluation of the impact of assistive artificial intelligence on ultrasound scanning for regional anaesthesia

BACKGROUND

Ultrasound-guided regional anaesthesia relies on the visualisation of key landmark, target, and safety structures on ultrasound. However, this can be challenging, particularly for inexperienced practitioners. Artificial intelligence (AI) is increasingly being applied to medical image interpretation, including ultrasound. In this exploratory study, we evaluated ultrasound scanning performance by non-experts in ultrasound-guided regional anaesthesia, with and without the use of an assistive AI device.

METHODS

Twenty-one anaesthetists, all non-experts in ultrasound-guided regional anaesthesia, underwent a standardised teaching session in ultrasound scanning for six peripheral nerve blocks. All then performed a scan for each block; half of the scans were performed with AI assistance and half without. Experts assessed acquisition of the correct block view and correct identification of sono-anatomical structures on each view. Participants reported scan confidence, experts provided a global rating score of scan performance, and scans were timed.

RESULTS

Experts assessed 126 ultrasound scans. Participants acquired the correct block view in 56/62 (90.3%) scans with the device compared with 47/62 (75.1%) without (P=0.031, two data points lost). Correct identification of sono-anatomical structures on the view was 188/212 (88.8%) with the device compared with 161/208 (77.4%) without (P=0.002). There was no significant overall difference in participant confidence, expert global performance score, or scan time.

CONCLUSIONS

Use of an assistive AI device was associated with improved ultrasound image acquisition and interpretation. Such technology holds potential to augment performance of ultrasound scanning for regional anaesthesia by non-experts, potentially expanding patient access to these techniques.

CLINICAL TRIAL REGISTRATION

NCT05156099.

Artificial intelligence in diagnostic ultrasonography

Artificial intelligence (AI) continues to change paradigms in the field of medicine with new applications that are applicable to daily life. The field of ultrasonography, which has been developing since the 1950s and continues to be one of the most powerful tools in the field of diagnosis, is also the subject of AI studies, despite its unique problems. It is predicted that many operations, such as appropriate diagnostic tool selection, use of the most relevant parameters, improvement of low-quality images, automatic lesion detection and diagnosis from the image, and classification of pathologies, will be performed using AI tools in the near future. Especially with the use of convolutional neural networks, successful results can be obtained for lesion detection, segmentation, and classification from images. In this review, relevant developments are summarized based on the literature, and examples of the tools used in the field are presented.

Artificial intelligence-based iliofemoral deep venous thrombosis detection using a clinical approach

Early diagnosis of deep venous thrombosis is essential for reducing complications, such as recurrent pulmonary embolism and venous thromboembolism. There are numerous studies on enhancing efficiency of computer-aided diagnosis, but clinical diagnostic approaches have never been considered. In this study, we evaluated the performance of an artificial intelligence (AI) algorithm in the detection of iliofemoral deep venous thrombosis on computed tomography angiography of the lower extremities to investigate the effectiveness of using the clinical approach during the feature extraction process of the AI algorithm. To investigate the effectiveness of the proposed method, we created synthesized images to consider practical diagnostic procedures and applied them to the convolutional neural network-based RetinaNet model. We compared and analyzed the performances based on the model's backbone and data. The performance of the model was as follows: ResNet50: sensitivity = 0.843 (± 0.037), false positives per image = 0.608 (± 0.139); ResNet152 backbone: sensitivity = 0.839 (± 0.031), false positives per image = 0.503 (± 0.079). The results demonstrated the effectiveness of the suggested method in using computed tomography angiography of the lower extremities, and improving the reporting efficiency of the critical iliofemoral deep venous thrombosis cases.

Acoustic Attenuation and Dispersion in Fatty Tissues and Tissue Phantoms Influencing Ultrasound Biomedical Imaging

The development of ultrasonic imaging techniques is optimized using artificial tissue phantoms before the practical applications. However, due to the strong attenuation and dispersion, accumulated fatty tissues can significantly impact the resolution and even feasibility of certain ultrasonic imaging modalities. An appropriate characterization of the acoustic properties on fatty phantoms can help the community to overcome the limitations. Some of the existing methods heavily overestimate attenuation coefficients by including the reflection loss and dispersion effects. Hence, in this study, we use numerical simulation-based comparison between two major attenuation measurement configurations. We further pointed out the pulse dispersion in viscoelastic tissue phantoms by simulations, which barely attracted attention in the existing studies. Using the selected attenuation and dispersion testing methods that were selected from the numerical simulation, we experimentally characterized the acoustic properties of common fatty tissue phantoms and compared the acoustic properties with the natural porcine fatty tissue samples. Furthermore, we selected one of the tissue phantoms to construct ultrasound imaging samples with some biomasses. With the known attenuation and dispersion of the tissue phantom, we showed the clarity enhancement of ultrasound imaging by signal post-processing to weaken the attenuation and dispersion effects.

The Use of Machine Learning Techniques to Predict Deep Vein Thrombosis in Rehabilitation Inpatients

BACKGROUND

Rehabilitation is crucial to recovering patients' dysfunction, improving their life quality, and promoting an early return to their family and society. In China, most patients in rehabilitation units are patients transferred from neurology, neurosurgery, and orthopedics, and most of these patients face problems such as continuously bedridden or varying degrees of limb dysfunction, all of which are risk factors for deep venous thrombosis. The formation of deep venous thrombosis can delay the recovery process and result in significant morbidity, mortality, and higher healthcare costs, so early detection and individualized treatment are needed. Machine learning algorithms can help develop more precise prognostic models, which can be of great significance in the development of rehabilitation training programs. In this study, we aimed to develop a model of deep venous thrombosis for inpatients in the Department of Rehabilitation Medicine at the Affiliated Hospital of Nantong University using machine learning methods.

METHODS

We analyzed and compared 801 patients in the Department of Rehabilitation Medicine using machine learning. Support vector machine, logistic regression, decision tree, random forest classifier, and artificial neural network were used to build models.

RESULTS

Artificial neural network was the better predictor than other traditional machine learnings. D-dimer levels, bedridden time, Barthel Index, and fibrinogen degradation products were common predictors of adverse outcomes in these models.

CONCLUSIONS

Risk stratification can help healthcare practitioners to achieve improvements in clinical efficiency and specify appropriate rehabilitation training programs.

Deep vein thrombosis - the role of ultrasound in the diagnosis and follow-up of patients

Deep vein thrombosis still represents a challenge regarding the diagnostics, treatment and follow-up. All this steps are often performed in the internal medicine ambulatory centers and such clinics therefore need to be aable to manage the whole process. Its key part is vascular sonography, which is needed to establish the diagnosis, the form of thrombosis and proper treatment course. There are two types of vascular sonography - Point of Care a expert sonography, being performed in two different regimes (diagnostic and follow up sonography). There are different demands for each of these two types, and each type is performed by physicians with different level of expertise. There are well defined criteria for performance a conclusions made for each type, and their precise fulfillment leads to establishment of diagnosis of DVT, early treatment initiation and setting of optimal strategy following the baseline treatment period.

Measuring the Compression Force Required for Vascular Shortening in Ultrasonic Vascular Models

Background Vascular compression is important for deep vein thrombosis screening. However, pressure analysis of ultrasound vessel models has not been performed. Therefore, we compared the human popliteal vein and several ultrasound vessel models at 50% compression. Methodology Four major ultrasound vascular models used in Japan and the popliteal vein of one subject constituted our measurement targets. Using a pressure-sensitive measuring device, the compressive force required to shorten the vessel diameter by 50% was determined. Results The compression force that shortened the popliteal vein by 50% was measured to be 191 ± 65 g. The blue phantom, ultrasound CV Pad II, ultrasound training block, and UGP-GEL required compression force of 701 ± 8 g, 265 ± 12 g, 697 ± 20 g, and 745 ± 15 g, respectively. The compression force for the ultrasound training block was 2.6 times higher than that for the ultrasound CV Pad II. The gel material around the vessels was the same; however, different vascular tubes required 2.6 times higher compression force. Conclusions This study showed that the objective numerical values of the compressive force were required to compress an ultrasound vascular model. Reproduction of the compressibility of veins required either removing the vascular structure or using thin tubing material.

The application of an age adjusted D-dimer threshold to rule out suspected venous thromboembolism (VTE) in an emergency department setting: a retrospective diagnostic cohort study

BACKGROUND

Venous Thromboembolic disease (VTE) poses a diagnostic challenge for clinicians in acute care. Over reliance on reference standard investigations can lead to over treatment and potential harm. We sought to evaluate the pragmatic performance and implications of using an age adjusted D-dimer (AADD) strategy to rule out VTE in patients with suspected disease attending an emergency department (ED) setting. We aimed to determine diagnostic test characteristics and assess whether this strategy would result in proportional imaging reduction and potential cost savings.

METHODS

Design: Single centre retrospective diagnostic cohort study. All patients > 50 years old evaluated for possible VTE who presented to the emergency department over a consecutive 12-month period between January and December 2016 with a positive D-dimer result. Clinical assessment records and reference standard imaging results were followed up by multiple independent adjudicators and coded as VTE positive or negative.

RESULTS

During the study period, there were 2132 positive D-dimer results. One thousand two hundred thirty-six patients received reference standard investigations. A total increase of 314/1236 (25.1%) results would have been coded as true negatives as opposed to false positive if the AADD cut off point had been applied, with 314 reference standard tests subsequently avoided. The AADD cut off had comparable sensitivity to the current cut off despite this increase in specificity; sensitivities for the diagnosis of DVT were 99.28% (95% CI 96.06-99.98%) and 97.72% for PE (95% CI 91.94% to 97.72). There were 3 false negative results using the AADD strategy.

CONCLUSIONS

In patients with suspected VTE with a low or moderate pre-test probability, the application of AADD appears to increase the proportion of patients in which VTE can be excluded without the need for reference standard imaging. This management strategy is likely to be associated with substantial reduction in anticoagulation treatment, investigations and cost/time savings.

Accurate assessment of the lung sliding artefact on lung ultrasonography using a deep learning approach

Pneumothorax is a potentially life-threatening condition that can be rapidly and accurately assessed via the lung sliding artefact generated using lung ultrasound (LUS). Access to LUS is challenged by user dependence and shortage of training. Image classification using deep learning methods can automate interpretation in LUS and has not been thoroughly studied for lung sliding. Using a labelled LUS dataset from 2 academic hospitals, clinical B-mode (also known as brightness or two-dimensional mode) videos featuring both presence and absence of lung sliding were transformed into motion (M) mode images. These images were subsequently used to train a deep neural network binary classifier that was evaluated using a holdout set comprising 15% of the total data. Grad-CAM explanations were examined. Our binary classifier using the EfficientNetB0 architecture was trained using 2535 LUS clips from 614 patients. When evaluated on a test set of data uninvolved in training (540 clips from 124 patients), the model performed with a sensitivity of 93.5%, specificity of 87.3% and an area under the receiver operating characteristic curve (AUC) of 0.973. Grad-CAM explanations confirmed the model's focus on relevant regions on M-mode images. Our solution accurately distinguishes between the presence and absence of lung sliding artefacts on LUS.

Investigation of two different human d-dimer assays in the horse

Background

D-dimer has value as a marker of thrombosis in critically ill horses and can provide additional information about prognosis. However, there are currently no equine species-specific d-dimer assays available, nor has there been any formal investigation of the applicability of human d-dimer assays in horses, so it is unknown, which assay performs best in this species. The aim of this study was therefore to evaluate and compare two human d-dimer assays for their applicability in horses.

The study included four groups of horses: clinically healthy horses, horses with gastrointestinal (GI) disease and mild systemic inflammation based on low serum amyloid A (SAA) (low SAA group), horses with GI disease and strong systemic inflammation based on high SAA (high SAA group) and, horses with thrombotic GI disease caused by Strongylus vulgaris (also called non-strangulating intestinal infarction (NSII)) (NSII group). The assays evaluated were the STAGO STA-Liatest D-di + (Stago) and NycoCard™ D-dimer (NycoCard). Intra- and inter-coefficients of variation (CV) were assessed on two d-dimer concentrations, and linearity under dilution was evaluated. A group comparison was performed for both assays across the four groups of horses. A Spaghetti plot, Spearman Correlation, Passing Bablok regression and Bland–Altman plot were used to compare methods in terms of agreement.

Results

Ten horses were included in the clinically healthy group, eight in the low SAA group, eight in the high SAA group, and seven in the NSII group. For the Stago assay, intra- and inter-CVs were below the accepted level except for one inter-CV. The NycoCard assay did not meet the accepted level for any of the CVs. The linearity under dilution was acceptable for both the Stago and NycoCard. In the group comparison, both methods detected a significantly higher d-dimer concentration in the high SAA and NSII groups compared to the clinically healthy group. Method agreement showed slightly higher d-dimer concentrations with NycoCard compared to Stago. The overall agreement was stronger for the lower d-dimer concentrations.

Conclusion

Both the Stago and the NycoCard were found to be applicable for use in horses but were not directly comparable.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12917-022-03313-5.

Evaluation of Machine Learning Algorithms for Early Diagnosis of Deep Venous Thrombosis

Deep venous thrombosis (DVT) is a disease that must be diagnosed quickly, as it can trigger the death of patients. Nowadays, one can find different ways to determine it, including clinical scoring, D-dimer, ultrasonography, etc. Recently, scientists have focused efforts on using machine learning (ML) and neural networks for disease diagnosis, progressively increasing the accuracy and efficacy. Patients with suspected DVT have no apparent symptoms. Using pattern recognition techniques, aiding good timely diagnosis, as well as well-trained ML models help to make good decisions and validation. The aim of this paper is to propose several ML models for a more efficient and reliable DVT diagnosis through its implementation on an edge device for the development of instruments that are smart, portable, reliable, and cost-effective. The dataset was obtained from a state-of-the-art article. It is divided into 85% for training and cross-validation and 15% for testing. The input data in this study are the Wells criteria, the patient’s age, and the patient’s gender. The output data correspond to the patient’s diagnosis. This study includes the evaluation of several classifiers such as Decision Trees (DT), Extra Trees (ET), K-Nearest Neighbor (KNN), Multi-Layer Perceptron Neural Network (MLP-NN), Random Forest (RF), and Support Vector Machine (SVM). Finally, the implementation of these ML models on a high-performance embedded system is proposed to develop an intelligent system for early DVT diagnosis. It is reliable, portable, open source, and low cost. The performance of different ML algorithms was evaluated, where KNN achieved the highest accuracy of 90.4% and specificity of 80.66% implemented on personal computer (PC) and Raspberry Pi 4 (RPi4). The accuracy of all trained models on PC and Raspberry Pi 4 is greater than 85%, while the area under the curve (AUC) values are between 0.81 and 0.86. In conclusion, as compared to traditional methods, the best ML classifiers are effective at predicting DVT in an early and efficient manner.