mHealth for image-based diagnostics of acute burns in resource-poor settings: studies on the role of experts and the accuracy of their assessments

Photographic Evaluation of Burn Depth via Telemedicine: Insights from Iranian Surgeons

Background

The accurate assessment of burn depth is crucial for determining appropriate treatment. Telemedicine has emerged as a promising tool for supporting burn diagnosis and decision-making, as it allows for remote consultation with burn specialists and access to high-quality imaging. The aim of this study was to evaluate the diagnostic capabilities of telemedicine in diagnosing burn depth.

Methods

A total of 15 Iranian surgeons participated in this study; they were presented with 13 images of partial thickness burn ulcers located on the extremities and trunk of patients. The participating surgeons were required to provide their diagnoses of burn type and depth, as well as the necessity of surgical intervention, and their responses were recorded.

Results

Data from 11 participants and 143 responses were analyzed. The average diagnostic accuracy for superficial burns was 79.3%, while for deep burns, it was 13.72%. The mean total diagnostic accuracy was 75.2%.

Conclusion

The results of this study suggest that photographs can be a reliable diagnostic tool for evaluating superficial burns. However, photographs are neither valid nor reliable for assessing burn depth. These findings have important implications for the use of telemedicine in burn diagnosis and indicate that additional diagnostic tools may be necessary for accurate assessment of deep burns.

Development and evaluation of deep learning algorithms for assessment of acute burns and the need for surgery

Assessment of burn extent and depth are critical and require very specialized diagnosis. Automated image-based algorithms could assist in performing wound detection and classification. We aimed to develop two deep-learning algorithms that respectively identify burns, and classify whether they require surgery. An additional aim assessed the performances in different Fitzpatrick skin types. Annotated burn (n = 1105) and background (n = 536) images were collected. Using a commercially available platform for deep learning algorithms, two models were trained and validated on 70% of the images and tested on the remaining 30%. Accuracy was measured for each image using the percentage of wound area correctly identified and F1 scores for the wound identifier; and area under the receiver operating characteristic (AUC) curve, sensitivity, and specificity for the wound classifier. The wound identifier algorithm detected an average of 87.2% of the wound areas accurately in the test set. For the wound classifier algorithm, the AUC was 0.885. The wound identifier algorithm was more accurate in patients with darker skin types; the wound classifier was more accurate in patients with lighter skin types. To conclude, image-based algorithms can support the assessment of acute burns with relatively good accuracy although larger and different datasets are needed.

Is the quality of mobile health applications for burns being adequately assessed?

High quality mobile health applications (mhealth apps) have the potential to enhance the prevention, diagnosis and treatment of burns. The primary aim of this study was to evaluate whether the quality of mhealth apps for burns care is being adequately assessed. The secondary aim was to determine whether these apps meet regulatory standards in the UK. We searched AMED, BNI, CINAHL, Cochrane library, Embase, Emcare, Medline and PsychInfo to identify studies assessing mhealth app quality for burns. The PRISMA reporting guideline was adhered to. Two independent reviewers screened abstracts to identify relevant studies. The quality of identified studies was assessed according to the framework proposed by Nouri et al, including design, information/content, usability, functionality, ethical issues, security/privacy and user-perceived value. Of the 28 included studies, none assessed all seven domains of quality. Design was assessed in 4/28 studies; information/content in 26/28 studies; usability in 12/28 studies; functionality in 10/28 studies; ethical issues were never assessed in any studies; security/privacy was not assessed; subjective assessment was made in 9/28 studies. 17/28 studies included apps that met the definition of 'medical device' according to MHRA guidance, yet only one app was appropriately certified with the UK Conformity Assessed (UKCA) mark. The quality of mHealth apps for burns are not being adequately assessed. The majority of apps should be considered medical devices according to UK standards, yet only one was appropriately certified. Regulatory bodies should support mhealth app developers, so as to improve quality control whilst simultaneously fostering innovation.

mHealth for image-based diagnostics of acute burns in resource-poor settings: studies on the role of experts and the accuracy of their assessments

Diagnostic assistance using mobile technology is instrumental to timely and adequate care in resource-scarce settings, particularly for acute burns. Little is known, however, as regards to how remote diagnostic consultation in burns affects the work process. This article reviews a doctoral thesis on this topic based on four studies conducted in the Western Cape, South Africa prior to and in a very early phase of the implementation of an app for burn remote diagnostic assistance. The aim was to increase knowledge on how remote diagnostic assistance for burn injuries can influence the role and work of medical experts in a resource-poor setting. The congruence model was used as a reference framework to study the ‘input’ (study 1), ‘tasks’ (studies 2 and 3) and ‘people’ (study 4) involved. The results show higher burn incidence in young children (75.4 per 10 000) and gender differences primarily among adults. The quality of images was considered by experts as better when viewed on smartphones and tablets than on computers. The accuracy of burn size assessments was high overall but low for burn depth (ICC = 0.82 and 0.53 respectively). Experts described four positions pertaining to remote consultations: clinical specialist, gatekeeper, mentor and educator. They perceived images as improving accuracy of consultation and stressed the need for verbal communication among clinicians during critical situations. In conclusion, experts are satisfied with the quality of images seen on handheld devices and can accurately assess burn size using these, yet burn depth assessment is more challenging without additional clinical information. mHealth for diagnostic assistance can benefit current image-based consultation by systematising information quality, introducing enhanced security and improved access to experts. Remaining challenges include the necessity of verbal communication in some instances and replacing existing informal organisational practices.

Adherence to the emergency management of severe burns referral criteria in burn patients admitted to a hospital with or without a specialized burn center

BACKGROUND

The primary aim was to determine to what extent referral and admission of burn patients to a hospital with or without a burn center was in line with the EMSB referral criteria.

METHODS

This was a retrospective, multicenter cohort study. Burn patients admitted from 2014 to 2018 to a hospital in the Southwest Netherland trauma region and Network Emergency Care Brabant were included in this study. Outcome measures were the adherence to the EMSB referral criteria.

RESULTS

A total of 1790 patients were included, of whom 951 patients were primarily presented to a non-burn center. Of these patients, 666 (70.0%) were managed according to the referral criteria; 263 (27.7%) were appropriately not referred, 403 (42.4%) were appropriately referred. Twenty (2.1%) were overtransferred, and 265 (27.9%) undertransferred. In 1213 patients treated at a burn center 1119 (92.3%) met the referral criteria. Adherence was lowest for electrical (N = 4; 14.3%) and chemical burns (N = 16; 42.1%), and was highest in 'children ≥5% total body surface area (TBSA) burned' (N = 109; 83.2%).

CONCLUSION

The overall adherence to the referral criteria of patients presented to a non-burn center was fairly high. However, approximately 25% was not transferred to a burn center while meeting the criteria. Most improvement for individual criteria can be achieved in patients with electrical and chemical burns.