Machine Learning Applications in the Evaluation and Management of Psoriasis: A Systematic Review

Background

Machine learning (ML), a subset of artificial intelligence (AI) that aims to teach machines to automatically learn tasks by inferring patterns from data, holds significant promise to aid psoriasis care. Applications include evaluation of skin images for screening and diagnosis as well as clinical management including treatment and complication prediction.

Objective

To summarize literature on ML applications to psoriasis evaluation and management and to discuss challenges and opportunities for future advances.

Methods

We searched MEDLINE, Google Scholar, ACM Digital Library, and IEEE Xplore for peer-reviewed publications published in English through December 1, 2019. Our search queries identified publications with any of the 10 computing-related keywords and "psoriasis" in the title and/or abstract.

Results

Thirty-three studies were identified. Articles were organized by topic and synthesized as evaluation- or management-focused articles covering 5 content categories: (A) Evaluation using skin images: (1) identification and differential diagnosis of psoriasis lesions, (2) lesion segmentation, and (3) lesion severity and area scoring; (B) clinical management: (1) prediction of complications and (2) treatment.

Conclusion

Machine learning has significant potential to aid psoriasis evaluation and management. Current topics popular in ML research on psoriasis are the evaluation of medical images, prediction of complications, and treatment discovery. For patients to derive the greatest benefit from ML advancements, it is helpful for dermatologists to have an understanding of ML and how it can effectively aid their assessments and decision-making.

Computer-aided analysis with Image J for quantitatively assessing psoriatic lesion area

BACKGROUND/AIMS

Body surface area is important in determining the severity of psoriasis. However, objective, reliable, and practical method is still in need for this purpose. We performed a computer image analysis (CIA) of psoriatic area using the image J freeware to determine whether this method could be used for objective evaluation of psoriatic area.

METHODS

Fifteen psoriasis patients were randomized to be treated with adalimumab or placebo in a clinical trial. At each visit, the psoriasis area of each body site was estimated by two physicians (E-method), and standard photographs were taken. The psoriasis area in the pictures was assessed with CIA using semi-automatic threshold selection (T-method), or manual selection (M-method, gold standard). The results assessed by the three methods were analyzed with reliability and affecting factors evaluated.

RESULTS

Both T- and E-method correlated strongly with M-method, and T-method had a slightly stronger correlation with M-method. Both T- and E-methods had a good consistency between the evaluators. All the three methods were able to detect the change in the psoriatic area after treatment, while the E-method tends to overestimate.

CONCLUSION

The CIA with image J freeware is reliable and practicable in quantitatively assessing the lesional of psoriasis area.

Psoriasis disease severity measures: comparing efficacy of treatments for severe psoriasis

Measurement of psoriasis disease severity and effectiveness of treatment involves both objective and subjective assessments.1 Comparing the efficacy of different treatments is complicated by the use of different metrics for measuring outcomes.2 Because these measures are not used routinely in clinical practice, interpreting these data, in particular assessing the degree of clinically meaningful improvement, is difficult. The drug approval process and product labeling reflect historical changes in standards of efficacy measurement.3 This paper reviews the metrics used to evaluate psoriasis treatment and compares available information on approved treatments for severe psoriasis. It further attempts to elucidate the value of these metrics and provide some guidance in properly evaluating the relative efficacy of current proven therapy with new treatments. While clinical trials are somewhat artificial, they provide proof that a drug is more effective than placebo. Efficacy in clinical practice, however, may be very different from the clinical trial setting. Comparison of efficacy under the current circumstances of varying evaluative metrics scales is possible with proper knowledge of the functionality of these methods.

Computerized quantification of psoriasis lesions with colour calibration: preliminary results

An evaluation was made of a fully automated index of psoriasis, termed Computer-assisted Area and Severity Index (CASI). This method requires taking digital photographs of the target skin area(s) with a colour reference marker, Casmatch. The CASI evaluates the severity of the psoriasis from the size and redness of the lesion(s). In five patients with mild psoriasis vulgaris mainly observed on their trunk, 18 photographs of the trunk were taken every 2 weeks. Three of the five patients [Psoriasis Area and Severity Index (PASI) of 3.0, 3.6 and 10.1, respectively] were treated with oral cyclosporin 3 mg/kg/day for 4 weeks. The mean +/- SD area of lesion selected by a dermatologist was 2.3 +/- 1.3% of the total skin area. This method achieved extraction performance for psoriasis of 72.1 +/- 19.4% for sensitivity and 97.4 +/- 2.0% for specificity. CASI correlated strongly with PASI (r = 0.92), but not with Skindex16 (r = 0.35). Although only erythema was evaluated, our preliminary results indicate that this method is capable of quantifying psoriasis lesions.

Method of assessing involved facial areas: rule of fours

BACKGROUND

Disease extent or affected area is probably the most easily and most frequently measured index of patient disability in dermatological disorders such as psoriasis. However, no standard method is available for assessing involved facial areas.

OBJECTIVES

To devise a method for assessing involved facial areas and to examine its validity and reliability.

METHODS

In study 1, the facial areas of 26 healthy adults were analysed and a new method for assessing facial areas was devised. In study 2, the validity and reliability of the devised method were investigated by 10 dermatologists using 30 photographs of patients with facial psoriasis.

RESULTS

In study 1, facial regions were defined to have percentage areas that were multiples of 4% of the total facial area (which we refer to as the rule of fours), i.e. forehead 24%, one cheek 20%, perioral area 8%, one aspect of an ear 4%, one periorbital area 4%, and nose 4%. In study 2, a comparison of values obtained using the rule of fours with those obtained using an image analysis system showed excellent intrarater and inter-rater reliability when using the rule of fours.

CONCLUSIONS

The rule of fours is useful for assessing involved facial areas. The application of this rule was found to provide an accurate and reliable method as compared with image analysis.

Computer-aided measurement of psoriatic lesion area in a multicenter clinical trial—Comparison to physician's estimations

BACKGROUND

There is a great need for accurate and informative outcome measures in clinical trials. In psoriasis, therapeutic effect is usually assessed by physician's estimation of psoriatic area and other signs. These measures were found to be un-reproducible. Several automated procedure for more reproducible measurement of psoriatic area were developed, but were not suitable for large-scale trials.

OBJECTIVES

In a multicenter clinical trial, we tested a method where the advantage of accurate computerized measurement of the area on the digital photograph was combined with physician's proficiency.

METHODS

The patients with psoriasis in four study centers were included in a placebo controlled clinical trial. They were examined and photographed before and after the therapy with calcipotriol ointment or placebo. The psoriatic area was manually outlined on the patient's photographs and the area was automatically measured by a computer. The areas estimated by physician and measured by computer were compared.

RESULTS

We found that computer-aided measurement of psoriatic lesion area improved the power of the clinical trial, compared to the standard approach, where physician's estimations of the psoriatic lesion area tend to overestimate. We also found that adapted PASI index, where the psoriatic area was not converted into an area grade, but was maintained as a continuous variable, also improved the power of the clinical trial.

CONCLUSIONS

Computer-aided measurement and an adapted PASI can be used as a powerful and reliable measure to evaluate the effects of antipsoriatic treatment.

Objective assessment of involved surface area in patients with psoriasis

BACKGROUND/AIMS

Although difficult, it is essential to assess the disease severity of psoriasis objectively. The visual grading method currently used for assessing the involved surface area in patients with psoriasis, which is included on the Psoriasis Area and Severity Index (PASI) scoring system, is a subjective method. The aim of this study was to establish the reliable objective method, in estimating the involved surface area in patients with psoriasis.

METHODS

Thirty patients with psoriasis participated in this study. The skin lesions of each patient were estimated using the visual grading method and an image analysis system.

RESULTS

The differences between five dermatologists' estimates of the involved surface area obtained from the visual grading method were highly significant (P<0.05, Kruskal-Wallis test). The comparison between the visual grading method and the image analysis system showed that the involved areas measured by the visual grading method were significantly higher than those from the image analysis (P<0.05, Wilcoxon signed rank test).

CONCLUSIONS

The visual grading method was neither objective nor reproducible. On the other hand, the image analysis system can be used as an objective method of measuring the involved surface area of patients with psoriasis. Therefore, the image analysis system can be used as a standard for the assessment of the involved area of patients with psoriasis.

Optical non-invasive approaches to diagnosis of skin diseases

A number of noninvasive approaches have been developed over the years to provide objective evaluation of the skin both in health and in disease. The advent of computers, as well as of lasers and photonics, has made it possible to develop additional techniques that were impossible a few years ago. These approaches provide the dermatologist with sensitive tools to measure the skin's condition in terms of physiologic parameters (e.g., color, erythema and pigmentation, induration, sebaceous and stratum corneum lipids, barrier function, etc.). Yet, a typical dermatologic diagnosis relies primarily on the trained eyes of the physician and to a lesser extent on information from other senses, such as touch and smell. The trained senses of the dermatologist backed by his/her brain form a powerful set of tools for evaluating the skin. The golden rule in diagnosis remains the histologic examination of a skin biopsy, a rather invasive method. These tools have served the profession well. The advent of ever faster and cheaper computers and of sensitive, inexpensive optical instrumentation of minimal dimensions provides the professional with the possibility of making objective measures of a number of skin parameters.

Clinical evaluation of a more rapid and sensitive Psoriasis Assessment Severity Score (PASS), and its comparison with the classic method of Psoriasis Area and Severity Index (PASI), before and after climatotherapy at the Dead-Sea

BACKGROUND

The Psoriasis Area and Severity Index (PASI) is used to quantify the extent of the disease, and to evaluate its improvement with treatment. It is considered to be a slow, rough, nonsensitive, and complex tool, with high interobserver variability and low reproducibility.

OBJECTIVES AND METHODS

To develop a simpler, more sensitive, and more rapid end-point determination for evaluating the psoriatic condition, and to compare its sensitivity with that of the classic PASI score in psoriatic patients undergoing 4-week climatotherapy at the Dead-Sea (Israel).

RESULTS

This study describes a new, rapid, and simple Psoriasis Assessment Severity Score (PASS), whose readings are spread over a longer scale, making the test more sensitive than PASI, and allow better differentiation.

CONCLUSIONS

The comparison between the classic PASI and our new PASS emphasizes the weight of the "sensitivity to change" (responsivity) in selecting a better evaluation method for psoriatic patients.