The availability of US services in underserved areas is limited by many factors. A lack of trained sonographers/sonologists and appropriate equipment, along with inadequate infrastructure, are two major ones.3 To help solve these challenges, telemedicine, including teleradiology and tele-ultrasound (TUS), can be employed to provide responses to clinical situations in the absence of on-site specialists.4 Indeed, multiple studies have shown that TUS improves access to, and reduces the costs of, healthcare in remote geographic regions with limited infrastructure.3,5
Indeed, with the growing availability of commercial off-the-shelf software, high-definition images using Windows or Android configurations can now be transferred around the world in seconds over Wi-Fi and 5G networks that provide 100 times faster data transmission rates (up to 10 gigabytes per second), and capacities more than 100 times that of 4G technology.
One drone-based product involves flying US units to inaccessible locations for catastrophic emergencies, and to vulnerable populations for whom transport is impractical or undesirable. This can be particularly useful in war zones, wilderness rescues, and other areas and circumstances with little access to imaging.
In obstetric US, a novel, self-operated home system that attaches to a smartphone (known as INSTINCT) is being used for fetal assessment, including a basic biophysical profile. In one study, it has demonstrated high accuracy in assessing fetal heart rate, amniotic fluid index, and fetal tone.7
Amid the limited resources and lack of expert sonographers in many LMICs, artificial intelligence (AI) offers the opportunity for even the inexperienced to acquire diagnostic-quality images. In the US and Europe, some agencies have been granted regulatory approval to process images using AI guidance. In this process, after entering anthropometric information, a transducer is placed on the patient. The operator is instructed on probe placement and movement, and the picture is automatically captured as soon as the device detects a high-quality image, essentially performing volume sweep imaging. AI-assisted TUS is still investigational; to ensure accuracy, a large database of condition-specific and normal images will be required to build the algorithms needed to train the technology. Nevertheless, AI-assisted guidance for all US procedures may eventually become commonplace. Deep-learning algorithms for precision US interpretation are also in early development; these tools are being studied for their ability to evaluate condition and injury severity for real-time clinical decision making.8
Tele-ultrasound has been shown to improve access to, and reduce the costs of healthcare in, remote areas with limited infrastructure. Rapidly evolving technology will promote increased implementation of TUS, especially in settings with few trained professionals. Affordable pocket/hand-held devices are increasingly available globally, providing excellent platforms for the application of TUS. As global connectivity expands, particularly with improved cellular and internet access, TUS is quickly becoming more feasible worldwide.
Although TUS applications are unlikely to fully replace hands-on scanning or human professional judgment in clinical decision making, they will continue to advance in the years ahead. Radiologists must remain up to date on technical innovations and consider leveraging TUS as a tool to expand high levels of care to those in need around the world.
The recent thrombectomy study from Brazil in low income population from public hospitals showed benefit for individuals with large vessel occlusion stroke presenting in the late time window (8-24 hours) and identified as suitable for treatment without requiring expensive imaging techniques. The study was coordinated by Hospital de Clnicas de Porto Alegre and sponsored by the Ministry of Health of Brazil.
The RESILIENT-Extend trial is the first major study of thrombectomy in the late time window conducted outside first-world countries, and shows the procedure also has benefit in a lower socioeconomic status population without the need for costly imaging equipment as MRI and CT perfusion.
However, there were some caveats that need to be considered, in particular a lack of benefit with thrombectomy in older patients (over 68 years of age) which can reflect the particular population enrolled in this study. The study may have identified frailty as a factor that correlates with reduced or lack of benefit of thrombectomy.
Raul Nogueira, who is professor of neurology and neurosurgery at the University of Pittsburgh, US, and Sheila Martins, who is a professor of neurology at Hospital de Clinicas Porto Alegre, Brazil and President of the World Stroke Organization, co- principal investigators of the study, presented the RESILIENT-Extend results at the International Stroke Conference 2024, being held in Phoenix, Arizona.
The lack of available advanced imaging techniques is a major challenge for implementing endovascular therapy in an extended time window especially in lower income countries and the main objective of study was to see if we could remove the need for advanced imaging to select patients with large vessel occlusion stroke in the late time window (8-24 hours) for thrombectomy. In this way our trial overlaps somewhat with the MR CLEAN-LATE Trial conducted in the Netherlands, although the two trials were conducted in very different socioeconomic populations.
The RESILIENT-Extend trial was conducted in the public health service of Brazil, coordinated by Hospital de Clnicas de Porto Alegre, and involved a different population of people than have been included in other thrombectomy trials which have mostly been conducted in first world countries. The public health system in Brazil is not well resourced and tends to care for patients at lower socioeconomic levels. These patients are fundamentally different from the average patients in the first world recruited into most other thrombectomy trials.
The trial enrolled 245 patients with a large vessel occlusion stroke within 8-24 hours of last known well. Patients were included who had a mismatch between the clinical severity as shown by the NIHSS score and the stroke burden on imaging as measured by the ASPECTS scores. They had to have relatively high NIHSS scores (8 or more) showing more severe strokes but also a high ASPECTS score (5-10) excluding patients with large areas of ischemic brain. There was also a sliding scale, created by Raul, that adjusted for age to avoid enrolling elderly patients with large strokes. These patients were identified exclusively using non contrast CT and CT angiography imaging.
795a8134c1