Imaging Scenarios Research Articles

Hyper‐Sampling Imaging by Measurement of Intra‐Pixel Quantum Efficiency Using Steady Wave Field

AbstractThe transition from optical film to digital image sensors (DIS) in imaging systems has brought great convenience to human life. However, the sampling resolution of DIS is considerably lower than that of optical film due to the limitation that the pixels are significantly larger than the silver halide molecules. How to break DIS's sampling limit and achieve high‐resolution imaging is highly desired for imaging applications. In the research, a novel mechanism that allows for a significant reduction in the smallest sampling unit of DIS to as small as 1/16th of a pixel, or even smaller, through measuring the intra‐pixel quantum efficiency for the first time and recomputing the image — a technique referred to as hyper‐sampling imaging (HSI) is developed. Employing the HSI method, the physical sampling resolution of regular DIS can be enhanced by 4 × 4 times or potentially higher, and detailed object information can be acquired. The HSI method has undergone rigorous testing in real‐world imaging scenarios, demonstrating its robustness and efficiency in overcoming the sampling constraints of conventional DIS. This advancement is particularly beneficial for applications such as remote sensing, long‐range reconnaissance, and astronomical observations, where the ability to capture fine details is paramount.

Acoustic sub-wavelength imaging via a virtual super-lens

Overcoming the diffraction limit has been a long-lasting pursuit for researchers owing to the great potential it offers in going beyond the fundamental resolution restriction in imaging processes. In acoustics, meta-lenses have been a promising way to achieve sub-wavelength imaging, the practical application of which, however, has been limited by expensive material manufacturing, complex system setup, and material loss. Here, we propose a set of procedures equivalent to a virtual super-lens that selectively amplifies the evanescent wave components in the measured acoustic field spectrum, thereby enabling super-resolution imaging without any auxiliary setups or purposely designed super-lens. The proposed virtual super-lens is experimentally verified by considering the imaging of an irregularly shaped sample with sub-wavelength features. We further demonstrate the robustness of the high-quality imaging performance remains acceptable with some environment background noises. In the light of the simple experimental setup involved, our proposed method is flexible and can be readily applied to various practical imaging scenarios.

Expert consensus on workflow of PET/CT with long axial field-of-view.

Positron emission tomography/computed tomography (PET/CT) imaging has been widely used in clinical practice. Long axial field-of-view (LAFOV) systems have enhanced clinical practice by leveraging their technological advantages and have emerged as the new state-of-the-art PET imaging modalities. A consensus was conducted to explore expert views in this emerging field to comprehensively elucidate the proposed workflow in LAFOV PET/CT examinations and highlight the potential challenges inherent in the workflow. A multidisciplinary task group formed by 28 experts from six countries over the world discussed and approved the consensus based on the published guidelines, peer-reviewed articles of LAFOV PET/CT, and the collective experience from clinical practice. This consensus focuses on the workflow that allows for a broader range of imaging protocols of LAFOV PET/CT, catering to diverse patient needs and in line with precision medicine principles. This consensus describes the workflows and imaging protocols of LAFOV PET/CT for various imaging scenarios including routine static imaging, dynamic imaging, low-activity imaging, fast imaging, prolonged imaging, delayed imaging, and dual-tracer imaging. In addition, imaging reconstruction and reviewing specific to LAFOV PET/CT imaging, as well as the main challenges facing installation and application of LAFOV PET/CT scanner were also summarized. This consensus summarized the various imaging workflow, imaging protocol, and challenges of LAFOV PET/CT imaging, aiming to enhance the clinical and research applications of these scanners.

Real-time location of acupuncture points based on anatomical landmarks and pose estimation models

IntroductionPrecise identification of acupuncture points (acupoints) is essential for effective treatment, but manual location by untrained individuals can often lack accuracy and consistency. This study proposes two approaches that use artificial intelligence (AI) specifically computer vision to automatically and accurately identify acupoints on the face and hand in real-time, enhancing both precision and accessibility in acupuncture practices.MethodsThe first approach applies a real-time landmark detection system to locate 38 specific acupoints on the face and hand by translating anatomical landmarks from image data into acupoint coordinates. The second approach uses a convolutional neural network (CNN) specifically optimized for pose estimation to detect five key acupoints on the arm and hand (LI11, LI10, TE5, TE3, LI4), drawing on constrained medical imaging data for training. To validate these methods, we compared the predicted acupoint locations with those annotated by experts.ResultsBoth approaches demonstrated high accuracy, with mean localization errors of less than 5 mm when compared to expert annotations. The landmark detection system successfully mapped multiple acupoints across the face and hand even in complex imaging scenarios. The data-driven approach accurately detected five arm and hand acupoints with a mean Average Precision (mAP) of 0.99 at OKS 50%.DiscussionThese AI-driven methods establish a solid foundation for the automated localization of acupoints, enhancing both self-guided and professional acupuncture practices. By enabling precise, real-time localization of acupoints, these technologies could improve the accuracy of treatments, facilitate self-training, and increase the accessibility of acupuncture. Future developments could expand these models to include additional acupoints and incorporate them into intuitive applications for broader use.

Imaging system high dynamic range colorimetric calibration method based on a digital chain

Research on high dynamic range (HDR) color management imposes critical requirements on calibration methods between imaging systems and standard radiation. This paper proposes a colorimetric calibration method based on digital chain measurement for imaging systems. First, a HDR colorimetric calibration process model for imaging systems is constructed based on an imaging chain. It includes a light source, target reflectance, optical system parameters, spectral sensitivities, and color matching functions. Subsequently, visual tristimulus values and three-channel response values of an imaging system are obtained using the proposed model in response to the same target, and the target characteristic parameters are adjusted to simulate different HDR imaging scenarios. Following that, various regression algorithms can be employed for HDR colorimetric calibration of imaging systems. The experimental findings demonstrate that the method proposed in this paper boasts a broader dynamic range and denser sampling, thereby enhancing the accuracy of colorimetric characterization models and achieving superior resolution in color measurement.

Lessons learned: strategies for implementing and the ongoing use of LI-RADS in your practice.

The establishment of the Liver Imaging Reporting and Data System (LI-RADS) in 2011 provided a comprehensive approach to standardized imaging, interpretation, and reporting of liver observations in patients diagnosed with or at risk for hepatocellular carcinoma (HCC). Each set of algorithms provides criteria pertinent to the various components of HCC management including surveillance, diagnosis, staging, and treatment response supported by a detailed lexicon of terms applicable to a wide range of liver imaging scenarios. Before its widespread adoption, the variability in the terminology of diagnostic criteria and definitions of imaging features led to significant challenges in patient management and made it difficult to replicate findings or apply them consistently. The integration of LI-RADS into the clinical setting has enhanced the efficiency and clarity of communication between radiologists, referring providers, and patients by employing a uniform language that averts miscommunications. LI-RADS has been strengthened with its integration into the American Association for Study of Liver Diseases practice guidelines. We will provide the background on the initial development of LI-RADS and reasons for development to serve as a starting point for conveying the system's benefits and evolution over the years. We will also suggest strategies for the implementation and maintenance of a LI-RADS program will be discussed.

Polarization-Enhanced Underwater Laser Range-Gated Imaging for Subaquatic Applications.

Laser range-gated underwater imaging technology, by removing most of the backscattering noise, can effectively increase image contrast and extend the detection range. The optical signal captured by a range-gated imaging system primarily comprises reflected light from the object and backscattered light from the surrounding water. Consequently, surfaces with low reflectivity or highly turbid water environments substantially constrain the applicability of the range-gated imaging system. To enhance the detection capability of underwater laser range-gated imaging, this paper proposes the incorporation of underwater polarized light imaging technology as an enhancement method. Based on polarization differences, backscattered light and reflected light from an object can be distinguished. Experimental results indicate that, compared to images obtained using a conventional range-gated laser imaging system, those captured with a polarization-enhanced system exhibit an increase of up to 47% for the corresponding Enhancement Measure Evaluation (EME) index. The proposed approach, which integrates polarization imaging with range-gated laser imaging, has the potential to broaden the applicability of underwater laser imaging scenarios, such as deep-sea exploration and military applications.

Diffraction tomography for incident Herglotz waves

Abstract Diffraction tomography is an inverse scattering technique used to reconstruct the spatial distribution of the material properties of a weakly scattering object. The object is exposed to radiation, typically light or ultrasound, and the scattered waves induced from different incident field angles are recorded. In conventional diffraction tomography, the incident wave is assumed to be a monochromatic plane wave, an unrealistic simplification in practical imaging scenarios. In this article, we extend conventional diffraction tomography by introducing the concept of customized illumination scenarios, with a pronounced emphasis on imaging with focused beams. More specifically, we consider incident Herglotz waves and extend the classical Fourier diffraction theorem to this setting. This yields a new two-step reconstruction process which we comprehensively evaluate through numerical experiments.

Radiation image reconstruction and uncertainty quantification using a Gaussian process prior

We propose a complete framework for Bayesian image reconstruction and uncertainty quantification based on a Gaussian process prior (GPP) to overcome limitations of maximum likelihood expectation maximization (ML-EM) image reconstruction algorithm. The prior distribution is constructed with a zero-mean Gaussian process (GP) with a choice of a covariance function, and a link function is used to map the Gaussian process to an image. Unlike many other maximum a posteriori approaches, our method offers highly interpretable hyperparamters that are selected automatically with the empirical Bayes method. Furthermore, the GP covariance function can be modified to incorporate a priori structural priors, enabling multi-modality imaging or contextual data fusion. Lastly, we illustrate that our approach lends itself to Bayesian uncertainty quantification techniques, such as the preconditioned Crank–Nicolson method and the Laplace approximation. The proposed framework is general and can be employed in most radiation image reconstruction problems, and we demonstrate it with simulated free-moving single detector radiation source imaging scenarios. We compare the reconstruction results from GPP and ML-EM, and show that the proposed method can significantly improve the image quality over ML-EM, all the while providing greater understanding of the source distribution via the uncertainty quantification capability. Furthermore, significant improvement of the image quality by incorporating a structural prior is illustrated.

Proof-of-concept of real-time electromagnetic guidance for gynecologic interstitial catheters in high dose rate brachytherapy

Background and PurposeThe addition of interstitial needles to intracavitary gynecologic (GYN) high dose rate (HDR) brachytherapy has been shown to improve target coverage and organs-at-risk (OAR) sparing. However, no commercial solution allows real-time guidance of interstitial catheter placement. This phantom study aimed to evaluate the feasibility of an electromagnetic (EM) tracking system guidance workflow for GYN HDR brachytherapy treatment in a magnetic resonance imaging (MRI) and real-time transrectal ultrasound (TRUS) fusion scenario. Materials and MethodsA clinical investigational system combining a treatment planning system and the EM tracking technology was used. The 3D T2 weighted magnetic resonance (MR) image set of a patient treated with intracavitary and interstitial HDR brachytherapy was retrospectively chosen. The MR image set was used to delineate the target and the OARs. A preplan was generated to determine needles positions in advance. The implant was reproduced in a water phantom. A 3D TRUS scan was acquired, and a rigid registration between the MR and the TRUS images was performed. ResultsThe accuracy of the EM tracking system was < 1 mm for both the sagittal and the transverse modes of the TRUS probe. Contours that were delineated on the MRI were propagated on the TRUS images after the rigid registration. Needle insertion was successfully guided in real time with the EM tracking system on the TRUS live image using the MRI contours for guidance. ConclusionBased on this proof-of-concept, real-time EM-guidance of interstitial needle for GYN HDR brachytherapy appears to be feasible.

Communicating Risk in Imaging: A Scoping Review of Risk Presentation in Patient Decision Aids

Best practices exist for communicating medical information to patients, however there is less emphasis on methods to communicate risks, especially in medical imaging. We conducted a scoping review of patient decision aids in medical imaging and characterized the presentation methods of imaging risks. Embase, Medline, CINAHL, and PsychINFO were searched to identify studies involving patient decision aids utilized in diagnostic imaging that communicated the risks. Study characteristics included the number and types of risks included, as well as the presentation type and how the probability of risks were communicated. The final study included 46 articles encompassing 27 distinct patient decision aids. Mammography was the most common imaging scenario (22/46), lung cancer screening (18/46), traumatic brain injury (5/46), and urolithiasis (1/46). All patient decision aids included risks associated with imaging, however the number of risk types varied from 2 to 9 (mean: 4, SD: 2). We identified 12 risks across the 27 decision aids, however no single study included all risks. Overall, most risks (65%) were communicated with text, and the presentation mode varied by type of risk. False positive risks were most commonly communicated using a visual format while radiation risk was most commonly communicated using text format. There was no consistent manner of communicating risk to patients and visual methods such as icon arrays were not consistently used. The variability of both included risks and the risk presentation modes in the PDAs may affect decision-making, especially in patients and caregivers with lower health literacy and numeracy.

MedLSAM: Localize and segment anything model for 3D CT images

Recent advancements in foundation models have shown significant potential in medical image analysis. However, there is still a gap in models specifically designed for medical image localization. To address this, we introduce MedLAM, a 3D medical foundation localization model that accurately identifies any anatomical part within the body using only a few template scans. MedLAM employs two self-supervision tasks: unified anatomical mapping (UAM) and multi-scale similarity (MSS) across a comprehensive dataset of 14,012 CT scans. Furthermore, we developed MedLSAM by integrating MedLAM with the Segment Anything Model (SAM). This innovative framework requires extreme point annotations across three directions on several templates to enable MedLAM to locate the target anatomical structure in the image, with SAM performing the segmentation. It significantly reduces the amount of manual annotation required by SAM in 3D medical imaging scenarios. We conducted extensive experiments on two 3D datasets covering 38 distinct organs. Our findings are twofold: (1) MedLAM can directly localize anatomical structures using just a few template scans, achieving performance comparable to fully supervised models; (2) MedLSAM closely matches the performance of SAM and its specialized medical adaptations with manual prompts, while minimizing the need for extensive point annotations across the entire dataset. Moreover, MedLAM has the potential to be seamlessly integrated with future 3D SAM models, paving the way for enhanced segmentation performance. Our code is public at https://github.com/openmedlab/MedLSAM.

Speckle autocorrelation separation method based on frequency intensity for tracking moving object beyond optical memory effect

Abstract Imaging objects behind a scattering medium beyond the optical memory effect (OME) range is vital in various imaging scenarios. Despite numerous proposed imaging methods aimed at separating objects, effectively tracking moving objects remains challenging. In this paper, we present a non-invasive scattering imaging method to reconstruct target beyond the OME range and track moving objects using a speckle separation and position estimation technique. A speckle autocorrelation separation method based on frequency intensity is proposed to eliminate the noise from the cross-correlation and to separate the objects from the mixed speckles. A position estimation technique is proposed to determine the positional relationships of the objects using speckle deconvolution using the point spread function of imaging regions. A non-invasive imaging system is designed to track the moving object without prior knowledge of the shape and distribution of objects. Experimental validation shows the efficiency of the method for separation and localization, allowing for the reconstruction of the target behind the scattering medium and tracking of the moving object behind the scattering medium. Moreover, the method is capable of separating and tracking a moving object when multiple objects are in the background.

Light-weight neural network for intra-voxel structure analysis.

We present a novel neural network-based method for analyzing intra-voxel structures, addressing critical challenges in diffusion-weighted MRI analysis for brain connectivity and development studies. The network architecture, called the Local Neighborhood Neural Network, is designed to use the spatial correlations of neighboring voxels for an enhanced inference while reducing parameter overhead. Our model exploits these relationships to improve the analysis of complex structures and noisy data environments. We adopt a self-supervised approach to address the lack of ground truth data, generating signals of voxel neighborhoods to integrate the training set. This eliminates the need for manual annotations and facilitates training under realistic conditions. Comparative analyses show that our method outperforms the constrained spherical deconvolution (CSD) method in quantitative and qualitative validations. Using phantom images that mimic in vivo data, our approach improves angular error, volume fraction estimation accuracy, and success rate. Furthermore, a qualitative comparison of the results in actual data shows a better spatial consistency of the proposed method in areas of real brain images. This approach demonstrates enhanced intra-voxel structure analysis capabilities and holds promise for broader application in various imaging scenarios.

AID-YOLO: An Efficient and Lightweight Network Method for Small Target Detector in Aerial Images

The progress of object detection technology is crucial for obtaining extensive scene information from aerial perspectives based on computer vision. However, aerial image detection presents many challenges, such as large image background sizes, small object sizes, and dense distributions. This research addresses the specific challenges relating to small object detection in aerial images and proposes an improved YOLOv8s-based detector named Aerial Images Detector-YOLO(AID-YOLO). Specifically, this study adopts the General Efficient Layer Aggregation Network (GELAN) from YOLOv9 as a reference and designs a four-branch skip-layer connection and split operation module Re-parameterization-Net with Cross-Stage Partial CSP and Efficient Layer Aggregation Networks (RepNCSPELAN4) to achieve a lightweight network while capturing richer feature information. To fuse multi-scale features and focus more on the target detection regions, a new multi-channel feature extraction module named Convolutional Block Attention Module with Two Convolutions Efficient Layer Aggregation Net-works (C2FCBAM) is designed in the neck part of the network. In addition, to reduce the sensitivity to position bias of small objects, a new function, Normalized Weighted Distance Complete Intersection over Union (NWD-CIoU_Loss) weight adaptive loss function, was designed in this study. We evaluate the proposed AID-YOLO method through ablation experiments and comparisons with other advanced models on the VEDAI (512, 1024) and DOTAv1.0 datasets. The results show that compared to the Yolov8s baseline model, AID-YOLO improves the mAP@0.5 metric by 7.36% on the VEDAI dataset. Simultaneously, the parameters are reduced by 31.7%, achieving a good balance between accuracy and parameter quantity. The Average Precision (AP) for small objects has improved by 8.9% compared to the baseline model (YOLOv8s), making it one of the top performers among all compared models. Furthermore, the FPS metric is also well-suited for real-time detection in aerial image scenarios. The AID-YOLO method also demonstrates excellent performance on infrared images in the VEDAI1024 (IR) dataset, with a 2.9% improvement in the mAP@0.5 metric. We further validate the superior detection and generalization performance of AID-YOLO in multi-modal and multi-task scenarios through comparisons with other methods on different resolution images, SODA-A and the DOTAv1.0 datasets. In summary, the results of this study confirm that the AID-YOLO method significantly improves model detection performance while maintaining a reduced number of parameters, making it applicable to practical engineering tasks in aerial image object detection.

Brainchop: Providing an Edge Ecosystem for Deployment of Neuroimaging Artificial Intelligence Models.

Deep learning has proven highly effective in various medical imaging scenarios, yet the lack of an efficient distribution platform hinders developers from sharing models with end-users. Here, we describe brainchop, a fully functional web application that allows users to apply deep learning models developed with Python to local neuroimaging data from within their browser. While training artificial intelligence models is computationally expensive, applying existing models to neuroimaging data can be very fast; brainchop harnesses the end user's graphics card such that brain extraction, tissue segmentation, and regional parcellation require only seconds and avoids privacy issues that impact cloud-based solutions. The integrated visualization allows users to validate the inferences, and includes tools to annotate and edit the resulting segmentations. Our pure JavaScript implementation includes optimized helper functions for conforming volumes and filtering connected components with minimal dependencies. Brainchop provides a simple mechanism for distributing models for additional image processing tasks, including registration and identification of abnormal tissue, including tumors, lesions and hyperintensities. We discuss considerations for other AI model developers to leverage this open-source resource.

Exploring the Combination of Computer Vision and Surgical Neuroanatomy: A Workflow Involving Artificial Intelligence for the Identification of Skull Base Foramina

The skull base is a complex region in neurosurgery, featuring numerous foramina. Accurate identification of these foramina is imperative to avoid intraoperative complications and to facilitate educational progress in neurosurgical trainees. The intricate landscape of the skull base often challenges both clinicians and learners, necessitating innovative identification solutions. We aimed to develop a computer vision model that automates the identification and labeling of the skull base foramina from various image formats, enhancing surgical planning and educational outcomes. We employed a deep learning methodology, specifically using a convolutional neural network architecture. Our model was trained on a dataset comprising of 3560 high-resolution, annotated images of the skull base, taken from various perspectives and lighting conditions to ensure model generalizability. Model performance was quantitatively assessed using precision and recall metrics. The convolutional neural network model demonstrated strong performance, achieving an average precision of 0.77. At a confidence threshold of 0.28, the model reached an optimal precision of 90.4% and a recall of 89.6%. Validation on an independent test set of images corroborated the model's capability to consistently and accurately identify and label multiple skull base foramina across diverse imaging scenarios. This study successfully introduces a highly accurate computer vision model tailored for the identification of skull base foramina, illustrating the model's potential as a transformative tool in anatomical education and intraoperative structure visualization. The findings suggest promising avenues for future research into automated anatomical recognition models, suggesting a trajectory toward increasingly sophisticated aids in neurosurgical operations and education.

A Chebyshev Confidence Guided Source-Free Domain Adaptation Framework for Medical Image Segmentation.

Source-free domain adaptation (SFDA) aims to adapt models trained on a labeled source domain to an unlabeled target domain without access to source data. In medical imaging scenarios, the practical significance of SFDA methods has been emphasized due to data heterogeneity and privacy concerns. Recent state-of-the-art SFDA methods primarily rely on self-training based on pseudo-labels (PLs). Unfortunately, the accuracy of PLs may deteriorate due to domain shift, thus limiting the effectiveness of the adaptation process. To address this issue, we propose a Chebyshev confidence guided SFDA framework to accurately assess the reliability of PLs and generate self-improving PLs for self-training. The Chebyshev confidence is estimated by calculating the probability lower bound of PL confidence, given the prediction and the corresponding uncertainty. Leveraging the Chebyshev confidence, we introduce two confidence-guided denoising methods: direct denoising and prototypical denoising. Additionally, we propose a novel teacher-student joint training scheme (TJTS) that incorporates a confidence weighting module to iteratively improve PLs' accuracy. The TJTS, in collaboration with the denoising methods, effectively prevents the propagation of noise and enhances the accuracy of PLs. Extensive experiments in diverse domain scenarios validate the effectiveness of our proposed framework and establish its superiority over state-of-the-art SFDA methods. Our paper contributes to the field of SFDA by providing a novel approach for precisely estimating the reliability of PLs and a framework for obtaining high-quality PLs, resulting in improved adaptation performance.

Domain adaptive noise reduction with iterative knowledge transfer and style generalization learning

Low-dose computed tomography (LDCT) denoising tasks face significant challenges in practical imaging scenarios. Supervised methods encounter difficulties in real-world scenarios as there are no paired data for training. Moreover, when applied to datasets with varying noise patterns, these methods may experience decreased performance owing to the domain gap. Conversely, unsupervised methods do not require paired data and can be directly trained on real-world data. However, they often exhibit inferior performance compared to supervised methods. To address this issue, it is necessary to leverage the strengths of these supervised and unsupervised methods. In this paper, we propose a novel domain adaptive noise reduction framework (DANRF), which integrates both knowledge transfer and style generalization learning to effectively tackle the domain gap problem. Specifically, an iterative knowledge transfer method with knowledge distillation is selected to train the target model using unlabeled target data and a pre-trained source model trained with paired simulation data. Meanwhile, we introduce the mean teacher mechanism to update the source model, enabling it to adapt to the target domain. Furthermore, an iterative style generalization learning process is also designed to enrich the style diversity of the training dataset. We evaluate the performance of our approach through experiments conducted on multi-source datasets. The results demonstrate the feasibility and effectiveness of our proposed DANRF model in multi-source LDCT image processing tasks. Given its hybrid nature, which combines the advantages of supervised and unsupervised learning, and its ability to bridge domain gaps, our approach is well-suited for improving practical low-dose CT imaging in clinical settings. Code for our proposed approach is publicly available at https://github.com/tyfeiii/DANRF.

SSP: self-supervised pertaining technique for classification of shoulder implants in x-ray medical images: a broad experimental study

Multiple pathologic conditions can lead to a diseased and symptomatic glenohumeral joint for which total shoulder arthroplasty (TSA) replacement may be indicated. The long-term survival of implants is limited. With the increasing incidence of joint replacement surgery, it can be anticipated that joint replacement revision surgery will become more common. It can be challenging at times to retrieve the manufacturer of the in situ implant. Therefore, certain systems facilitated by AI techniques such as deep learning (DL) can help correctly identify the implanted prosthesis. Correct identification of implants in revision surgery can help reduce perioperative complications and complications. DL was used in this study to categorise different implants based on X-ray images into four classes (as a first case study of the small dataset): Cofield, Depuy, Tornier, and Zimmer. Imbalanced and small public datasets for shoulder implants can lead to poor performance of DL model training. Most of the methods in the literature have adopted the idea of transfer learning (TL) from ImageNet models. This type of TL has been proven ineffective due to some concerns regarding the contrast between features learnt from natural images (ImageNet: colour images) and shoulder implants in X-ray images (greyscale images). To address that, a new TL approach (self-supervised pertaining (SSP)) is proposed to resolve the issue of small datasets. The SSP approach is based on training the DL models (ImageNet models) on a large number of unlabelled greyscale medical images in the domain to update the features. The models are then trained on a small labelled data set of X-ray images of shoulder implants. The SSP shows excellent results in five ImageNet models, including MobilNetV2, DarkNet19, Xception, InceptionResNetV2, and EfficientNet with precision of 96.69%, 95.45%, 98.76%, 98.35%, and 96.6%, respectively. Furthermore, it has been shown that different domains of TL (such as ImageNet) do not significantly affect the performance of shoulder implants in X-ray images. A lightweight model trained from scratch achieves 96.6% accuracy, which is similar to using standard ImageNet models. The features extracted by the DL models are used to train several ML classifiers that show outstanding performance by obtaining an accuracy of 99.20% with Xception+SVM. Finally, extended experimentation has been carried out to elucidate our approach’s real effectiveness in dealing with different medical imaging scenarios. Specifically, five different datasets are trained and tested with and without the proposed SSP, including the shoulder X-ray with an accuracy of 99.47% and CT brain stroke with an accuracy of 98.60%.