1. |
- Cirillo, Marco Domenico, et al.
(author)
-
Improving burn depth assessment for pediatric scalds by AI based on semantic segmentation of polarized light photography images
- 2021
-
In: Burns. - : Elsevier. - 0305-4179 .- 1879-1409. ; 47:7, s. 1586-1593
-
Journal article (peer-reviewed)abstract
- This paper illustrates the efficacy of an artificial intelligence (AI) (a convolutional neural network, based on the U-Net), for the burn-depth assessment using semantic segmentation of polarized high-performance light camera images of burn wounds. The proposed method is evaluated for paediatric scald injuries to differentiate four burn wound depths: superficial partial-thickness (healing in 0–7 days), superficial to intermediate partial-thickness (healing in 8–13 days), intermediate to deep partial-thickness (healing in 14–20 days), deep partial-thickness (healing after 21 days) and full-thickness burns, based on observed healing time.In total 100 burn images were acquired. Seventeen images contained all 4 burn depths and were used to train the network. Leave-one-out cross-validation reports were generated and an accuracy and dice coefficient average of almost 97% was then obtained. After that, the remaining 83 burn-wound images were evaluated using the different network during the cross-validation, achieving an accuracy and dice coefficient, both on average 92%.This technique offers an interesting new automated alternative for clinical decision support to assess and localize burn-depths in 2D digital images. Further training and improvement of the underlying algorithm by e.g., more images, seems feasible and thus promising for the future.
|
|
2. |
- Cirillo, Marco Domenico, et al.
(author)
-
Time-Independent Prediction of Burn Depth using Deep Convolutional Neural Networks
- 2019
-
In: Journal of Burn Care & Research. - : Oxford University Press. - 1559-047X .- 1559-0488. ; 40:6, s. 857-863
-
Journal article (peer-reviewed)abstract
- We present in this paper the application of deep convolutional neural networks, which are a state-of-the-art artificial intelligence (AI) approach in machine learning, for automated time-independent prediction of burn depth. Colour images of four types of burn depth injured in first few days, including normal skin and background, acquired by a TiVi camera were trained and tested with four pre-trained deep convolutional neural networks: VGG-16, GoogleNet, ResNet-50, and ResNet-101. In the end, the best 10-fold cross-validation results obtained from ResNet- 101 with an average, minimum, and maximum accuracy are 81.66%, 72.06% and 88.06%, respectively; and the average accuracy, sensitivity and specificity for the four different types of burn depth are 90.54%, 74.35% and 94.25%, respectively. The accuracy was compared to the clinical diagnosis obtained after the wound had healed. Hence, application of AI is very promising for prediction of burn depth and therefore can be a useful tool to help in guiding clinical decision and initial treatment of burn wounds.
|
|
3. |
- Elawa, Sherif, et al.
(author)
-
Microcirculatory changes in the skin after postmastectomy radiotherapy in women with breast cancer
- 2024
-
In: Scientific Reports. - : Nature Publishing Group. - 2045-2322. ; 14:1
-
Journal article (peer-reviewed)abstract
- Postmastectomy radiotherapy (PMRT) increases the risk for complications after breast reconstruction. The pathophysiological mechanism underlying this increased risk is not completely understood. The aim of this study was to examine if there is a relationship between PMRT and microvascular perfusion in the skin directly after, and at 2 and 6 months after PMRT and to assess if there is impaired responsiveness to a topically applied vasodilator (Methyl nicotinate-MN) after PMRT. Skin microvascular responses after PMRT were measured on two sites in the irradiated chest wall of 22 women before, immediately after, and at 2 and 6 months after unilateral PMRT with the contralateral breast as a control. A significant increase in basal skin perfusion was observed in the irradiated chest wall immediately after RT (p < 0.0001). At 2 and 6 months after RT, there was no longer a difference in basal skin perfusion compared to the contralateral breast and compared to baseline. Similarly, the blood flow response in the skin after application of MN was stronger immediately after RT compared to before RT (p < 0.0001), but there was no difference at later time points. These results indicate that the increased risk for complications after surgical procedures are not directly related to changes in skin perfusion and microvascular responsiveness observed after postmastectomy RT.
|
|
4. |
- Elmasry, Moustafa, 1981-, et al.
(author)
-
Laser speckle contrast imaging in children with scalds : Its influence on timing of intervention, duration of healing and care, and costs
- 2019
-
In: Burns. - : Elsevier. - 0305-4179 .- 1879-1409. ; 45:4, s. 798-804
-
Journal article (peer-reviewed)abstract
- BackgroundScalds are the most common type of burn injury in children, and the initial evaluation of burn depth is a problem. Early identification of deep dermal areas that need excision and grafting would save unnecessary visits and stays in hospital. Laser speckle contrast imaging (LSCI) shows promise for the evaluation of this type of burn. The aim of this study was to find out whether perfusion measured with LSCI has an influence on the decision for operation, duration of healing and care period, and costs, in children with scalds.MethodsWe studied a group of children with scalds whose wounds were evaluated with LSCI on day 3–4 after injury during the period 2012–2015. Regression (adjustment for percentage total body surface area burned (TBSA%), age, and sex) was used to analyse the significance of associations between degree of perfusion and clinical outcome.ResultsWe studied 33 children with a mean TBSA% of 6.0 (95% CI 4.4–7.7)%. Lower perfusion values were associated with operation (area under the receiver-operating characteristic curve 0.86, 95% CI 0.73–1.00). The perfusion cut-off with 100% specificity for not undergoing an operation was ≥191 PU units (66.7% sensitivity and 72.7% accurately classified). Multivariable analyses showed that perfusion was independently associated with duration of healing and care period.ConclusionLower perfusion values, as measured with LSCI, are associated with longer healing time and longer care period. By earlier identification of burns that will be operated, perfusion measurements may further decrease the duration of care of burns in children with scalds.
|
|
5. |
- Mirdell, Robin, 1989-
(author)
-
Blood Flow Dynamics in Burns
- 2019
-
Doctoral thesis (other academic/artistic)abstract
- Objectives:Burns of intermediate thickness are hard to evaluate clinically. This often leads to unnecessary delays of up to 14 days before a surgical decision can be made. To counter this, several objective methods have been developed to determine the healing potential of the wound. Over the years, measurement of perfusion has proven to be the most successful method for evaluation of healing potential. Laser Doppler imaging (LDI) is currently the most used method and can determine surgical need 2 days after injury with an accuracy >90%. There are however emerging techniques like laser speckle contrast imaging (LSCI), which also measure perfusion. LSCI have several advantages over LDI and is easier to use. LSCI can also investigate aspects of the microcirculation, previously not possible with LDI. The aim of this thesis was to investigate LSCI’s ability to evaluate surgical need in burns of indeterminate partial-thickness. The first objective was to investigate the dynamics of perfusion the first 14 days after injury. The purpose was to find the optimal time-window for perfusion measurements. The next goal was to determine the accuracy of different perfusion cut-offs. In this second study, the benefit of a subsequent measurement was also investigated. After this, interobserver variation between different profession groups was studied. Both the agreement of perfusion measurements and observer assessments were evaluated. Finally, cardiac vasomotion in combination with perfusion (pulsatility) was investigated as a method to determine surgical need <48 hours after injury. Methods:Perfusion was measured in a total of 77 patients at the Department of Plastic Surgery, Hand Surgery and Burns at Linköping University Hospital, Sweden. Most of these patients were children and the most common type of burn was scalds. A laser speckle contrast imager (PeriCam PSI System, Perimed AB, Järfälla, Sweden) was used to measure perfusion. Results: In the first paper we showed a clear relation between perfusion dynamics and the healing potential of the wound. The changes in perfusion were largest the first 5 days after injury, why this time interval was selected for subsequent papers. Perfusion measurements done day 3-4 after injury could predict surgical need with a sensitivity of 100% (95% CI: 83.9-100%) and a specificity of 90.4% (95% CI: 83.8-94.9%). If two measurements were used, <24 hours and 3-4 days after injury, the accuracy was 100%. Furthermore, we found that different observers could consistently predict perfusion, while there was a large variation in their clinical assessments. This was not improved by extensive burn experience. Finally, pulsatility could be used to predict surgical need the same day as the injury occurred with a sensitivity of 100% (95% CI: 88.1-100%) and a specificity of 98.8% (95% CI: 95.7- 99.9%). Conclusions: LSCI is a promising method for evaluation of burns and provides several benefits over LDI. The surgical need of burns can be determined mere hours after injury when pulsatility is measured. However, the benefits of early scald diagnostics in children with LSCI need to be evaluated in a prospective study before the method is ready for routine clinical use.
|
|
6. |
- Mirdell, Robin, 1989-, et al.
(author)
-
Data on microcirculatory perfusion dips in the resting nail bed
- 2018
-
In: Data in Brief. - : Elsevier. - 2352-3409. ; 21, s. 1232-1235
-
Journal article (peer-reviewed)abstract
- This article contains the raw data from the article entitled: "The presence of synchronized perfusion dips in the microcirculation of the resting nail bed" Mirdell et al. (in press). A laser speckle contrast imager (LSCI) was used to make a total of 21 recordings of the perfusion in the resting nail bed of 10 healthy test subjects. The first 10 recordings were acquired after 5?min of acclimatization. An additional 10 recordings were acquired in the same test subjects, after 20?min of acclimatization. In the last recording, a digital nerve block was applied to the left dig III. The data show the presence of highly irregular perfusion variations, a phenomenon we like to call perfusion dips. The data also show how the perfusion dips can be abolished through a digital nerve block. An algorithm for the quantification of the perfusion dips is included in the data.
|
|