Waail Mahmod Lafta has a total of 24 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Deep Transfer Learning with Enhanced Feature Fusion for Detection of Abnormalities in X-ray Images

    by Zaenab Alammar, Laith Alzubaidi, Jinglan Zhang, Yuefeng Li, Waail Lafta, Yuantong Gu

    Medical image classification poses significant challenges in real-world scenarios. One major obstacle is the scarcity of labelled training data, which hampers the performance of image-classification algorithms and generalisation. Gathering sufficient labelled data is often difficult and time-consuming in the medical domain, but deep learning (DL) has shown remarkable performance, although it typically requires a large amount of labelled data to achieve optimal results. Transfer learning (TL) has played a pivotal role in reducing the time, cost, and need for a large number of labelled images. This paper presents a novel TL approach that aims to overcome the limitations and disadvantages of TL that are characteristic of an ImageNet dataset, which belongs to a different domain. Our proposed TL approach involves training DL models on numerous medical images that are similar to the target dataset. These models were then fine-tuned using a small set of annotated medical images to leverage the knowledge gained from the pre-training phase. We specifically focused on medical X-ray imaging scenarios that involve the humerus and wrist from the musculoskeletal radiographs (MURA) dataset. Both of these tasks face significant challenges regarding accurate classification. The models trained with the proposed TL were used to extract features and were subsequently fused to train several machine learning (ML) classifiers. We combined these diverse features to represent various relevant characteristics in a comprehensive way. Through extensive evaluation, our proposed TL and feature-fusion approach using ML classifiers achieved remarkable results. For the classification of the humerus, we achieved an accuracy of 87.85%, an F1-score of 87.63%, and a Cohen’s Kappa coefficient of 75.69%. For wrist classification, our approach achieved an accuracy of 85.58%, an F1-score of 82.70%, and a Cohen’s Kappa coefficient of 70.46%. The results demonstrated that the models trained using our proposed TL approach outperformed those trained with ImageNet TL. We employed visualisation techniques to further validate these findings, including a gradient-based class activation heat map (Grad-CAM) and locally interpretable model-independent explanations (LIME). These visualisation tools provided additional evidence to support the superior accuracy of models trained with our proposed TL approach compared to those trained with ImageNet TL. Furthermore, our proposed TL approach exhibited greater robustness in various experiments compared to ImageNet TL. Importantly, the proposed TL approach and the feature-fusion technique are not limited to specific tasks. They can be applied to various medical image applications, thus extending their utility and potential impact. To demonstrate the concept of reusability, a computed tomography (CT) case was adopted. The results obtained from the proposed method showed improvements.


  • Open Access

    New Approach Measuring the Wheel/Rail Interaction Loads

    by Yuri P. Boronenko, Rustam V. Rahimov, Waail M. Lafta

    This chapter suggested new methods for monitoring the dynamic processes of rolling stock/rail interaction. This study develops a new technical solution for measuring the wheel/rail interaction forces on a significant part of the sleeper. The theoretical part of this study, using FEM, confirm the ability of piecewise continuous recording of vertical and lateral forces from the wheel/rail interaction by measuring the stresses in two sections of the rail. Also, the optimum location of strain gauges and the effective length of the measuring zone have been determined. The experimental part of this study has been carried out on the stands and the railway track to confirm the effectiveness of the method to determine the vertical and lateral wheel/rail interaction forces, increase the reliable statistical data, improve the measurement accuracy, reducing the time and cost compared with current testing methods. The developed method is recommended to determine the wheel/rail interaction forces and identify defects on the wheels when diagnosing rolling stock on operational and travel regimes.

  • Develop a New Approach Measuring the Wheel/Rail Interaction Loads

    by Yuriy Boronenko, Rustam Rahimov, Waail Mahmod Lafta

    Determine and control the impact of rolling stock on the railway track, one of the significant subjects of railway engineering, especially with heavy traffic and innovative freight cars with increased axle loads. Different methods utilized to measure the lateral impact of rolling stock on a railway based on the use of strain gauges installed on the rail differ in the location of strain gauges and the specifics of processing the received signals. The shortage of these methods that the lateral force arising from the wheel/rail interaction determined when the wheel position over the strain gauges sections. Therefore, continuous registration of details in the wheel/rail contact is impossible. Multiple passes of the test rolling stock along the measuring section are required to receive the right results. In this article, a new method developed to continue recording the lateral forces of the wheel/rail interaction by measuring stresses in two sections of the rail on a significant part of the sleeper space. The railway track experiments approved this method’s ability to restore the lateral force of not more than 4% standard deviation along the measuring zone’s length and increased the volume of reliable statistical data obtained, improved the measurement accuracy, and reduced the time and cost.


  • Continuous Monitoring of the Wheel-Rail Contact Vertical Forces by Using a Variable Measurement Scale

    by Yu. P. Boronenko, R. V. Rahimov, Waail M. Lafta, S. V. Dmitriev, A. V. Belyankin, D. A. Sergeev

    This paper presents a new technology of continuous measurement and recording of forces in wheel/rail contact by measuring strains in two rail cross-sections, which allows increasing the measurement accuracy. The method for processing the measurement data is proposed with a variable scale between strains and force, depending on the instantaneous position of the wheel with respect to a span between sleepers. The gage characteristics obtained by rolling a wheel over the rail between two sleepers are approximated by using Fourier series abilities and used for the assessment of wheel/rail forces. By using the suggested technology, the amount of information obtained increased 2 times.


  • Development of new polymer composite materials for the flooring of rail carriage

    by Zaynitdinova Olmos Irikovich, Rahimov Rustam Vyacheslavovich, Waail Mahmod Lafta, Ruzmetov Yadgor Ozodovich

    The subject of improving the competitiveness and efficiency of railway rolling stock is highly demanded as the railway transport consider the nerve of the goods transport in the hole word. One of the requirements to develop the railway transport related to the continuous improvement of existing materials and find a new material which would satisfy such conflicting requirements as relia-bility, sustainability, producibility and the manufacturing process automation possibility. The researches show that use of polymer composite materials (PCM) in the production of both passenger and freight railway rolling stock, provides lightweight, improves its performance, increases durability and reduces operating costs.


  • Sensitive and fast response graphite pressure sensor fabricated by a solvent-free approach

    by Toan Dinh, Hoang-Phuong Phan, Tuan-Khoa Nguyen, Jarred Fastier-Wooller, Abu Riduan Md Foisal, Waail Mahmod Lafta, Nam-Trung Nguyen, Dzung Viet Dao

    This work demonstrates a biodegradable resistive pressure sensor fabricated by a solvent-free, low cost and highly efficient method. The sensor shows a high sensitivity of 4.77 %/kPa, a fast response of 0.3 ms and a relaxation time of 8 ms. The fast response time, flexibility, and capability to be safely attached to the human body and measure the motion of the Unger in real-time indicate its potential for human motion detection, touch displays, and tactile sensing applications.