Dataset Information

Automated detection of actinic keratoses in clinical photographs.

ABSTRACT:

Background

Clinical diagnosis of actinic keratosis is known to have intra- and inter-observer variability, and there is currently no non-invasive and objective measure to diagnose these lesions.

Objective

The aim of this pilot study was to determine if automatically detecting and circumscribing actinic keratoses in clinical photographs is feasible.

Methods

Photographs of the face and dorsal forearms were acquired in 20 volunteers from two groups: the first with at least on actinic keratosis present on the face and each arm, the second with no actinic keratoses. The photographs were automatically analysed using colour space transforms and morphological features to detect erythema. The automated output was compared with a senior consultant dermatologist's assessment of the photographs, including the intra-observer variability. Performance was assessed by the correlation between total lesions detected by automated method and dermatologist, and whether the individual lesions detected were in the same location as the dermatologist identified lesions. Additionally, the ability to limit false positives was assessed by automatic assessment of the photographs from the no actinic keratosis group in comparison to the high actinic keratosis group.

Results

The correlation between the automatic and dermatologist counts was 0.62 on the face and 0.51 on the arms, compared to the dermatologist's intra-observer variation of 0.83 and 0.93 for the same. Sensitivity of automatic detection was 39.5% on the face, 53.1% on the arms. Positive predictive values were 13.9% on the face and 39.8% on the arms. Significantly more lesions (p<0.0001) were detected in the high actinic keratosis group compared to the no actinic keratosis group.

Conclusions

The proposed method was inferior to assessment by the dermatologist in terms of sensitivity and positive predictive value. However, this pilot study used only a single simple feature and was still able to achieve sensitivity of detection of 53.1% on the arms.This suggests that image analysis is a feasible avenue of investigation for overcoming variability in clinical assessment. Future studies should focus on more sophisticated features to improve sensitivity for actinic keratoses without erythema and limit false positives associated with the anatomical structures on the face.

SUBMITTER: Hames SC

PROVIDER: S-EPMC4304708 | biostudies-literature | 2015

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Automated detection of actinic keratoses in clinical photographs.

Hames Samuel C SC Sinnya Sudipta S Tan Jean-Marie JM Morze Conrad C Sahebian Azadeh A Soyer H Peter HP Prow Tarl W TW

PloS one 20150123 1

<h4>Background</h4>Clinical diagnosis of actinic keratosis is known to have intra- and inter-observer variability, and there is currently no non-invasive and objective measure to diagnose these lesions.<h4>Objective</h4>The aim of this pilot study was to determine if automatically detecting and circumscribing actinic keratoses in clinical photographs is feasible.<h4>Methods</h4>Photographs of the face and dorsal forearms were acquired in 20 volunteers from two groups: the first with at least on ...[more]

PMID: 25615930

Similar Datasets

Project description:Importance:Photodynamic therapy (PDT) is an effective and cosmetically favorable treatment modality for actinic keratoses (AKs). However, prolonged incubation times and pain associated with treatment are burdensome to the patient and a hindrance to widespread use of PDT as standard field therapy for AK. Objective:To evaluate efficacy and pain associated with microneedle expedited PDT. Design, Setting, and Participants:The Microneedle Photodynamic Therapy II (MNPDT-II) study was a randomized, single-blinded, split-face controlled, 2-arm clinical trial. Thirty-three participants with AK on the face were recruited in a university dermatology outpatient clinic from 2015 to 2016, and 32 participants completed the study. Interventions:Participants were randomized into 2 incubations arms, either 10-minute or 20-minute aminolevulinic acid (ALA) incubation times, after pretreatment with a microneedle roller (200 um) vs a sham roller. They were blinded to the laterality of microneedle and sham roller assignments. After incubation, they were exposed to blue light (Blu-U, Dusa Pharmaceuticals) for 1000 seconds for a total fluence of 10 J/cm2. Main Outcomes and Measures:The primary outcome was to quantitatively measure AK resolution, and the secondary outcome was to assess pain associated with microneedle pretreatment. Results:Thirty-three individuals were recruited and randomized to either the 20-minute or the 10-minute incubation arm. Thirty-two participants completed the study with a mean follow-up time of 34.5 days in the 20-minute group, and 30.2 days in the 10-minute group. For the 20-minute incubation arm, average AK clearance was 76% vs 58% on the sham side (P < .01), including 3 patients with complete clearance, although not statistically significant (P = .25). Pain assessment on the visual analog scale (VAS) during blue light illumination was not significantly different between the microneedle and sham sides (0.7 and 0.4; P = .28), respectively. For the 10-minute incubation arm AK clearance for the microneedle pretreated side was 43% compared with 38% on the sham side (P = .66). Pain during the blue light exposure was not significantly different between the microneedle and sham sides, 4.5 mm and 3.4 mm (P = .21), respectively. Conclusions and Relevance:Photodynamic therapy with microneedle pretreatment at a 20-minute ALA incubation time significantly improved AK clearance with efficacy similar to that of a conventional 1-hour ALA incubation time. The additional advantage to expedited treatment was that the procedure was virtually painless. However, expedited exposure of a 10-minute ALA incubation time did not reach significantly different AK clearance from the sham control. Trial Registration:clinicaltrials.gov Identifier: NCT02594644.

Project description:BackgroundThe risk of developing cutaneous squamous cell carcinoma (SCC) is markedly increased in organ transplant recipients (OTRs) compared to the normal population. Next to sun exposure, the immunosuppressive regimen is an important risk factor for the development of SCC in OTRs. Various gene mutations (e.g. TP53) and genetic alterations (e.g. loss of CDKN2A, amplification of RAS) have been found in SCCs. The aim of this genome-wide study was to identify pathways and genomic alterations that are consistently involved in the formation of SCCs and their precursor lesions, actinic keratoses (AKs).MethodsTo perform the analysis in an isogenic background, RNA and DNA were isolated from SCC, AK and normal (unexposed) epidermis (NS) from each of 13 OTRs. Samples were subjected to genome-wide expression analysis and genome SNP analysis using Illumina's HumanWG-6 BeadChips and Infinium II HumanHap550 Genotyping BeadChips, respectively. mRNA expression results were verified by quantitative PCR.ResultsHierarchical cluster analysis of mRNA expression profiles showed SCC, AK and NS samples to separate into three distinct groups. Several thousand genes were differentially expressed between epidermis, AK and SCC; most upregulated in SCCs were hyperproliferation related genes and stress markers, such as keratin 6 (KRT6), KRT16 and KRT17. Matching to oncogenic pathways revealed activation of downstream targets of RAS and cMYC in SCCs and of NFκB and TNF already in AKs. In contrast to what has been reported previously, genome-wide SNP analysis showed very few copy number variations in AKs and SCCs, and these variations had no apparent relationship with observed changes in mRNA expression profiles.ConclusionVast differences in gene expression profiles exist between SCC, AK and NS from immunosuppressed OTRs. Moreover, several pathways activated in SCCs were already activated in AKs, confirming the assumption that AKs are the precursor lesions of SCCs. Since the drastic changes in gene expression appeared unlinked to specific genomic gains or losses, the causal events driving SCC development require further investigation. Other molecular mechanisms, such as DNA methylation or miRNA alterations, may affect gene expression in SCCs of OTRs. Further study is required to identify the mechanisms of early activation of NFκB and TNF, and to establish whether these pathways offer a feasible target for preventive intervention among OTRs.

Dataset Information

Automated detection of actinic keratoses in clinical photographs.

Background

Objective

Methods

Results

Conclusions

Publications

Automated detection of actinic keratoses in clinical photographs.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets