Evaluating and assessing the prevalence of bedside ultrasound in emergency departments in China

Lightweight deep neural networks for cholelithiasis and cholecystitis detection by point-of-care ultrasound

BACKGROUND AND OBJECTIVE

Emergency physicians (EPs) frequently deal with abdominal pain, including that is caused by either gallstones or acute cholecystitis. Easy access and low cost justify point-of-care ultrasound (POCUS) use as a first-line test to detect these diseases; yet, the detection performance of POCUS by EPs is unreliable, causing misdiagnoses with serious impacts. This study aimed to develop a machine learning system to detect and localize gallstones and to detect acute cholecystitis by ultrasound (US) still images taken by physicians or technicians for preliminary diagnoses.

METHODS

Abdominal US images (> 89,000) were collected from 2386 patients in a hospital database. We constructed training sets for gallstones with or without cholecystitis (N = 10,971) and cholecystitis with or without gallstones (N = 7348) as positives. Validation sets were also constructed for gallstones (N = 2664) and cholecystitis (N = 1919). We applied a single-shot multibox detector (SSD) and a feature pyramid network (FPN) to classify and localize objects using image features extracted by ResNet-50 for gallstones, and MobileNet V2 to classify cholecystitis. The deep learning models were pretrained using the COCO-2017 and ILSVRC-2012 datasets.

RESULTS

Using the validation sets, the SSD-FPN-ResNet-50 and MobileNet V2 achieved areas under the receiver operating characteristics curve of 0.92 and 0.94, respectively. The inference speeds were 21 (47.6 frames per second, fps) and 7 ms (142.9 fps).

CONCLUSIONS

A machine learning system was developed to detect and localize gallstones, and to detect cholecystitis, with acceptable discrimination and speed. This is the first study to develop this system for either gallstone or cholecystitis detection with absence or presence of each one. After clinical trials, this system may be used to assist EPs, including those in remote areas, for detecting these diseases.

Point-of-care ultrasound in primary care: a systematic review of generalist performed point-of-care ultrasound in unselected populations

BACKGROUND

Both the interest and actual extent of use of point-of-care ultrasound, PoCUS, among general practitioners or family physicians are increasing and training is also increasingly implemented in residency programs. However, the amount of research within the field is still rather limited compared to what is seen within other specialties in which it has become more established, such as in the specialty of emergency medicine. An assumption is made that what is relevant for emergency medicine physicians and their populations is also relevant to the general practitioner, as both groups are generalists working in unselected populations. This systematic review aims to examine the extent of use and to identify clinical studies on the use of PoCUS by either general practitioners or emergency physicians on indications that are relevant for the former, both in their daily practice and in out-of-hours services.

METHODS

Systematic searches were done in PubMed/MEDLINE using terms related to general practice, emergency medicine, and ultrasound.

RESULTS

On the extent of use, we identified 19 articles, as well as 26 meta-analyses and 168 primary studies on the clinical use of PoCUS. We found variable, but generally low, use among general practitioners, while it seems to be thoroughly established in emergency medicine in North America, and increasingly also in the rest of the world. In terms of clinical studies, most were on diagnostic accuracy, and most organ systems were studied; the heart, lungs/thorax, vessels, abdominal and pelvic organs, obstetric ultrasound, the eye, soft tissue, and the musculoskeletal system. The studies found in general either high sensitivity or high specificity for the particular test studied, and in some cases high total accuracy and superiority to other established diagnostic imaging modalities. PoCUS also showed faster time to diagnosis and change in management in some studies.

CONCLUSION

Our review shows that generalists can, given a certain level of pre-test probability, safely use PoCUS in a wide range of clinical settings to aid diagnosis and better the care of their patients.