Musculoskeletal ultrasound (MSKUS) is a powerful, non-invasive imaging tool that has significantly transformed the diagnosis and management of rheumatologic conditions [1]. Its ability to provide real-time, high-resolution imaging of joints, tendons, ligaments, and soft tissues makes it invaluable for detecting early inflammation, guiding procedures, and monitoring disease progression [2]. MSKUS is radiation-free (unlike X-ray), cost-effective (unlike MRI), and allows for dynamic simultaneous assessments of multiple joint and tendons sites at the bed-side, enabling clinicians to evaluate joint movement and detect subtle abnormalities that static imaging might miss. Additionally, its role in guiding injections and aspirations enhances precision, procedural accuracy, and patient outcomes.

However, despite its advantages, several barriers limit the widespread use of ultrasound in rheumatology. These include the lack of standardized training, limited access to education, and the high cost of ultrasound equipment [3,4]. Addressing these challenges is essential to ensuring broader adoption and effective utilization of this technology.

A Handheld Ultrasound (HHUS) is a portable, compact imaging device that provides real-time scanning at the point of care. Unlike conventional cart-based ultrasound (CBU) machines, which are bulky and require dedicated space, HHUS are lightweight and connect to smartphones or tablets for image display. These devices are revolutionizing medical imaging by providing quick, accessible assessments at the point of care in various clinical settings. In rheumatology, HHUS are particularly valuable for evaluating musculoskeletal inflammation, guiding procedures, and improving early diagnosis and disease monitoring.

One of the major advantages of HHUS is its affordability, as it is substantially less expensive than conventional CBU machines. The advantages and disadvantages of HHUS are summarized in Table 1. As a result, adoption of HHUS devices in rheumatology is rapidly increasing. However, it is essential to recognize that HHUS devices are not identical to CBU regarding image quality, functionality, and integration capabilities. This review will compare the benefits and limitations of both methods and assess the accuracy of HHUS, drawing on both existing literature and the authors’ clinical experience. Given the limited data on HHUS in the musculoskeletal system, we will also include insights from other medical fields.

Several factors influence overall ultrasound image quality, some operator-dependent, while others are related to the device itself. One key device-related factor is the probe frequency. High-frequency probes offer superior resolution but at the expense of reduced penetration, making them ideal for imaging superficial structures. Advances in ultrasound technology have enabled the use of higher-frequency probes, allowing for detailed visualization of small joints, significantly enhancing diagnostic accuracy in musculoskeletal assessments. Traditionally, CBU systems have supported higher-frequency probes, offering an advantage in image quality over HHUS devices. Recent advancements in ultrasound technology have led to increasing availability of higher-frequency probes in HHUS devices, narrowing the gap in image quality and expanding their utility in musculoskeletal assessments. However, advanced ultrasound machines offer greater processing power and superior image enhancements algorithms enabling them to provide even clearer images. Therefore, there are inevitable differences in image quality, at least with today's technology, which may or may not impact the overall judgement in the clinical decision making.

One way to judge the image quality is the subjective assessment of experts in the field. There are a few studies that assessed the quality and characteristics of HHUS devices, by comparing various machines for certain characteristics, image quality being one of these. In the study by Le et al., four HHUS brands were compared by experts on various image characteristics, including detail resolution, penetration, clutter, contrast resolution, and overall satisfaction with the quality of the image [5]. In another cross-sectional study, 35 experts acquired three standard views using six HHUS brands, stating that the overall satisfaction between the devices did not differ statistically [6]. A non-inferiority trial compared eight HHUS brands for their B-image quality in the intensive care setting on a) good spatial resolution, b) grey scale contrast and c) overall image quality. In addition, the ultrasound videos were reassessed for how accurately the clinical question could be answered, by using a CBU device for comparison [7]. In another similar non-inferiority study aimed to evaluate B-scan quality and the possibility of answering clinical questions by comparing eight HHUS devices, performed by an ultrasound specialist [8]. Finally, in another prospective observational study, 12 ultrasound experts tested nine different HHUS devices on healthy subjects [9]. In the previously mentioned studies, it is not possible to determine which HHUS device performs best due to several important limitations. Notably, none of the devices were specifically tested for musculoskeletal ultrasound or for the evaluation of inflammatory arthritis. In addition, the assessments were conducted by non-rheumatologist operators, and the evaluations involved subjective assessments, which may introduce bias, making difficult a consistent comparison across devices. Importantly, the general idea across these studies was that no single devices was perceived as superior in all evaluated categories. While these studies have primarily relied on expert opinion and subjective scoring scales, there is also emerging evidence from objective technical evaluations. For instance, a German study conducted a technical assessment of five HHUS devices using a general-purpose phantom to measure image quality and software performance objectively. In addition to technical testing, clinical performance was evaluated by two expert gastroenterologists. Although variations were observed in the performance of the different handheld devices, these differences did not appear to impact their practical use in routine clinical activities significantly [10]. Across these studies, no single HHUS device consistently demonstrated superior image quality across evaluations. This variability likely reflects the inherently subjective nature of image quality assessments and differences in users' clinical needs, targeted anatomical regions, and intended applications. Notably, none of the existing studies were conducted specifically within rheumatology, limiting their findings' relevance to musculoskeletal applications. Image quality and usability may vary when applied in a rheumatology-specific context. Before broader implementation, we recommend evaluating any HHUS device within its intended clinical setting. A direct comparison with a CBU system in patients with inflammatory arthritis can provide valuable insight into the device's performance and suitability for routine practice, specific to the needs of rheumatologists.

When assessing ultrasound systems, ease of use represents an important factor. CBU devices are less portable and impractical for use in multiple settings, whereas HHUS devices offer the major advantage of easy portability. However, ease of use extends beyond just mobility. Device weight can impact usability, particularly for wireless models, which tend to be heavier due to built-in batteries. While battery-powered operation allows use in any setting, it requires regular charging. HHUS devices also have fewer adjustable settings, making them more user-friendly for novice users but potentially limiting flexibility in image optimization for experienced operators. Additionally, device heating can become a concern during prolonged scanning sessions, affecting both user comfort and performance. In addition, entering patient details and image labelling is less convenient in HHUS.

There are a few studies that aimed to compare different HHUS devices for their ease of use, according to expert opinion. In the study by Le et al., overall ease of use were rated, in addition to the physical characteristics, software navigability and maneuverability [5]. In another study by Perez-Sanchez, comparison was made for ease of use of the devices’ software, physical characteristics and maneuverability and for the overall satisfaction [6]. Other study reports some partial features of ease of use, such as in the Merkel et al. study, which included “handling” [9]. Similar to the image quality, findings were inconsistent across evaluations, likely due to the large variations across study settings/indications; and the devices were tested in non-musculoskeletal settings.

The wireless design of some HHUS devices offers greater portability and convenience. However, a major trade-off is increased weight, as the battery must be housed within the probe itself (in addition to the need for regular charging). Wired ultrasound probes typically weigh around 100 g, whereas wireless models can range from 300 to 400 g—a substantial increase. While this weight difference may not be an issue for short scans, it can become a concern during longer examinations, potentially affecting both the scanner's comfort and the quality of imaging due to unintended pressure on the scanned area.

Securely storing ultrasound images and videos is crucial for both clinical decision-making and medicolegal purposes. There are two primary storage options: 1) Images and scans can be stored directly on the smartphone, tablet, or computer connected to the ultrasound probe. 2) Cloud-based platforms, that allows access to images remotely. If the scans are stored on personal devices, it is very important that the security measures are in place, similar to hospital IT protocols, to protect patient data from breaches or unauthorized access. If Cloud-Based Platforms are used, ownership and who has access to the scans must be clarified. In hospital settings, privacy regulations may restrict the use of cloud-based storage due to concerns about data security and compliance with HIPAA and other regulatory frameworks.

HHUS devices can also be connected to electronic medical record (EMR) systems, but the integration depends on the device, software compatibility, and institutional infrastructure. Some devices have built-in software that allows direct upload of images and reports into EMR systems. Cloud storage solutions can also be linked to EMRs. Devices that support DICOM can transmit ultrasound images to a hospital's PACS, from which they can be accessed through the EMR. Since EMR integration capabilities vary across different devices and healthcare systems, institutions often collaborate with IT teams and device manufacturers to ensure compliance with data security regulations (e.g., HIPAA) and smooth workflow implementation.

As previously noted, differences in image quality between HHUS devices and CBU systems are unavoidable. Therefore, it is essential to understand whether these differences lead to any differences in the clinical decisions, which can be tested by agreement studies:a)

Reliability studies from non-musculoskeletal fields:

The agreement of both types of devices has mostly been investigated in cardiology. A recent systematic review of the literature on 42 studies- 6598 patients showed a median kappa of 0.72 to measure the ejection fraction by both devices [11]. Geers et al. assessed hospitalized patients with acute coronary syndrome, finding a good concordance for the same objective (kappa 0.60–1.00) [12]. Providença et al. reported in a systematic literature review about the implementation of screening for rheumatic heart diseases, that the handheld echocardiography has a high sensitivity, specificity and general high accuracy for the diagnosis of rheumatic cardiac disease, compared to standard echocardiography [13]. In respiratory medicine, some studies also show varying results on the accuracy of HHUS, but unlike other fields, the gold standard in respirology is usually other imaging methods. One study evaluated HHUS used by newly certified operators for diagnosing pneumonia and found low sensitivity compared to expert and CT-based diagnoses [14]. However, in another study that compared different ultrasound systems for specific US findings, there was good agreement in detecting B-lines, pleural effusion and consolidations [15]. Operator-related factors such as experience can significantly influence the level of agreement between ultrasound devices. In a comparative study that evaluates the non-inferiority of HHUS versus CBU using videos of lung pathologies, less experienced clinicians displayed significantly more variability in their assessments [16]. Of note, we found only one randomized clinical trial on the topic. This study was conducted by Gibbons et al. comparing HHUS and CBU in patients requiring scans of heart, lungs, kidneys, aorta or biliary system. Among 110 patients included, both systems demonstrated similar sensitivity, specificity, and overall diagnostic accuracy [17].b)

Handheld Ultrasound in Musculoskeletal Conditions:

Although evidence supporting HHUS in cardiac and respiratory conditions has grown, research in musculoskeletal diseases remains limited. Some studies have evaluated peripheral nerves and muscles due their accessibility, with contrasting results regarding agreement and accuracy compared to standard ultrasound. One study of 43 patients found that cross-sectional area measurements (CSA) of the median nerve at the carpal tunnel were comparable between the two devices [18]. In the same way, Alfuraih et al. also investigated the CSA of the median nerve in asymptomatic subjects, reporting excellent intraoperative reproducibility, which was impacted by the operator's experience [19]. Regarding the evaluation of muscle thickening, a cross-sectional study using “Ultrasound Sarcopenia Index” in elderly patients showed consistent muscle thickening evaluation with both HHUS and CBU [20]. In another study, where two novice evaluators were included and muscle thickening in healthy patients was evaluated with a portable device and a conventional machine, it showed excellent accuracy between devices [21]. Evaluating joint conditions in anatomically complex areas with HHUS remains uncertain, as could be seen in a 2020 study comparing both systems, HHUS and CBU, in assessing joints like shoulders, hands, hips, and knees, and found diagnostic agreement in only 65 % of cases. However, 96 % of these discrepancies were deemed clinically insignificant [22].

In Rheumatology, both B-mode and Doppler-mode ultrasonography are essential, with the latter playing a particularly important role due to its correlation with disease activity in inflammatory arthritis. While B-mode imaging can provide substantial information to guide clinical decision-making, excluding Doppler-mode assessment would neglect a critical component of ultrasound evaluation, especially in inflammatory conditions.

The importance of incorporating Doppler mode is underscored by a study conducted by Corte et al., which assessed the accuracy and performance of a HHUS compared to a CBU system in 32 patients with inflammatory arthritis (20 with rheumatoid arthritis, 10 with psoriatic arthritis, one with gout, and one with lupus-related arthritis), examining a total of 186 joints. The study reported an overall agreement in B-mode findings of 97 %, with a Cohen's kappa of 0.90, indicating excellent concordance. However, in Doppler mode, the portable device failed to detect any Doppler signal, whereas the conventional system identified Doppler activity in 61 joints [23].

Even when focusing solely on B-mode imaging, the level of agreement between HHUS and CBU systems remains uncertain. This was exemplified in a study by Murdoch et al., where two rheumatologists independently assessed six lower extremity sites in 21 participants using both a CBU and a HHUS device. Although the findings were broadly similar across devices, inter-device agreement at the anatomical site level was only fair to good. Specifically, kappa values were 0.22 for double contour sign, 0.46 for tophi, 0.63 for erosions, and 0.37 for aggregates [24].

The concordance between HHUS and CBU in evaluating enthesitis has been examined in a study conducted by our group. In this study, patients with psoriatic arthritis (PsA) and at least one tender and swollen joint underwent consecutive evaluations using both portable and conventional ultrasound devices, both using B mode and Doppler modalities. The findings revealed moderate to substantial agreement in the detection of elementary lesions of enthesitis, synovitis, nail lesions and erosions, between the HHUS and CBU as a gold standard. Of particular note, unlike other reports where Doppler data were absent or not analyzed, this study demonstrated a moderate-substantial concordance in inflammatory parameters when using Doppler mode, highlighting the potential of portable ultrasound when appropriately applied [25,26].

Assessing the clinical value of HHUS devices in inflammatory arthritis screening remains challenging, largely due to difficulties in defining appropriate and measurable outcomes. Most existing studies have focused on their utility in screening for inflammatory arthritis, more specifically PsA among patients with psoriasis.

A study by Grobelski et al. evaluated 140 psoriasis patients with arthralgia using a HHUS device operated by dermatologists trained in MSKUS. Up to six joints per patient were scanned, primarily in the hands and wrists. Prior to ultrasound, screening sensitivity and specificity for PsA were 88.2 % and 54.4 %, respectively. Following ultrasound, these shifted to 70.6 % and 90.4 %. Importantly, ultrasound findings revised clinical suspicion in 46 patients—PsA was ruled out in 45 of them—raising the positive predictive value from 25.4 % to 56.5 % [27]. In contrast, Bartsch et al. examined 40 psoriasis patients using an enthesis-focused US protocol performed by trained dermatologists with a HHUS device. A rheumatologist later conducted a comprehensive evaluation and repeated the imaging using a standard CBU system. While dermatologists were generally able to distinguish PsA from psoriasis, the study concluded that HHUS did not significantly improve diagnostic accuracy. The authors also cautioned against over-reliance on HHU devices, citing concerns about misinterpretation [28]. These contrasting results likely stem more from differences in anatomical targets—joint-based vs. enthesis-focused—than from device limitations alone.

Focusing solely on the operator's role—rather than image quality, resolution, or technological aspects of portable ultrasound devices—an important question arises: Can any tools help bridge the gap between expert and novice operators? To address this, several studies have explored the use of artificial intelligence (AI) to enhance diagnostic accuracy, with promising yet somewhat controversial findings.

Some studies support and encourage AI-assisted ultrasound, demonstrating its potential to improve accuracy. For instance, a study found that oncologists using HHUS portable ultrasound with AI assistance were able to automatically calculate left ventricular volume, and the accuracy of detecting an ejection fraction below 50 % was high [29]. Similarly, another study revealed that non-expert operators using AI-assisted ultrasound could effectively predict mortality and hospital readmission rates based on systolic function measurements [30]. Additionally, AI-supported HHUS enabled non-experts to acquire image quality comparable to that of expert echocardiographers in most cases [31].

However, not all studies support AI's reliability in ultrasound interpretation. One study, aimed at validating AI-assisted HHUS for left ventricular ejection fraction evaluation, found that the tool tended to underestimate left ventricular volumes, raising concerns about its accuracy [32].

Despite these mixed results, AI remains modifiable and trainable, meaning that continued research and validation could refine its accuracy. As AI technology evolves, it has the potential to narrow the gap between novice and expert operators, shorten the learning curve, and ultimately make ultrasound assessments more accessible and reliable in clinical practice.