Automated Machine Learning Analysis of Patients With Chronic Skin Disease Using a Medical Smartphone App: Retrospective Study.

BACKGROUND: Rapid digitalization in health care has led to the adoption of digital technologies; however, limited trust in internet-based health decisions and the need for technical personnel hinder the use of smartphones and machine learning applications. To address this, automated machine learning (AutoML) is a promising tool that can empower health care professionals to enhance the effectiveness of mobile health apps. OBJECTIVE: We used AutoML to analyze data from clinical studies involving patients with chronic hand and/or foot eczema or psoriasis vulgaris who used a smartphone monitoring app. The analysis focused on itching, pain, Dermatology Life Quality Index (DLQI) development, and app use. METHODS: After extensive data set preparation, which consisted of combining 3 primary data sets by extracting common features and by computing new features, a new pseudonymized secondary data set with a total of 368 patients was created. Next, multiple machine learning classification models were built during AutoML processing, with the most accurate models ultimately selected for further data set analysis. RESULTS: Itching development for 6 months was accurately modeled using the light gradient boosted trees classifier model (log loss: 0.9302 for validation, 1.0193 for cross-validation, and 0.9167 for holdout). Pain development for 6 months was assessed using the random forest classifier model (log loss: 1.1799 for validation, 1.1561 for cross-validation, and 1.0976 for holdout). Then, the random forest classifier model (log loss: 1.3670 for validation, 1.4354 for cross-validation, and 1.3974 for holdout) was used again to estimate the DLQI development for 6 months. Finally, app use was analyzed using an elastic net blender model (area under the curve: 0.6567 for validation, 0.6207 for cross-validation, and 0.7232 for holdout). Influential feature correlations were identified, including BMI, age, disease activity, DLQI, and Hospital Anxiety and Depression Scale-Anxiety scores at follow-up. App use increased with BMI >35, was less common in patients aged >47 years and those aged 23 to 31 years, and was more common in those with higher disease activity. A Hospital Anxiety and Depression Scale-Anxiety score >8 had a slightly positive effect on app use. CONCLUSIONS: This study provides valuable insights into the relationship between data characteristics and targeted outcomes in patients with chronic eczema or psoriasis, highlighting the potential of smartphone and AutoML techniques in improving chronic disease management and patient care.

Using Automated Machine Learning to Predict Necessary Upcoming Therapy Changes in Patients With Psoriasis Vulgaris and Psoriatic Arthritis and Uncover New Influences on Disease Progression: Retrospective Study.

BACKGROUND: Psoriasis vulgaris (PsV) and psoriatic arthritis (PsA) are complex, multifactorial diseases significantly impacting health and quality of life. Predicting treatment response and disease progression is crucial for optimizing therapeutic interventions, yet challenging. Automated machine learning (AutoML) technology shows promise for rapidly creating accurate predictive models based on patient features and treatment data. OBJECTIVE: This study aims to develop highly accurate machine learning (ML) models using AutoML to address key clinical questions for PsV and PsA patients, including predicting therapy changes, identifying reasons for therapy changes, and factors influencing skin lesion progression or an abnormal Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) score. METHODS: Clinical study data from 309 PsV and PsA patients were extensively prepared and analyzed using AutoML to build and select the most accurate predictive models for each variable of interest. RESULTS: Therapy change at 24 weeks follow-up was modeled using the extreme gradient boosted trees classifier with early stopping (area under the receiver operating characteristic curve [AUC] of 0.9078 and logarithmic loss [LogLoss] of 0.3955 for the holdout partition). Key influencing factors included the initial systemic therapeutic agent, the Classification Criteria for Psoriatic Arthritis score at baseline, and changes in quality of life. An average blender incorporating three models (gradient boosted trees classifier, ExtraTrees classifier, and Eureqa generalized additive model classifier) with an AUC of 0.8750 and LogLoss of 0.4603 was used to predict therapy changes for 2 hypothetical patients, highlighting the significance of these factors. Treatments such as methotrexate or specific biologicals showed a lower propensity for change. An average blender of a random forest classifier, an extreme gradient boosted trees classifier, and a Eureqa classifier (AUC of 0.9241 and LogLoss of 0.4498) was used to estimate PASI (Psoriasis Area and Severity Index) change after 24 weeks. Primary predictors included the initial PASI score, change in pruritus levels, and change in therapy. A lower initial PASI score and consistently low pruritus were associated with better outcomes. BASDAI classification at onset was analyzed using an average blender of a Eureqa generalized additive model classifier, an extreme gradient boosted trees classifier with early stopping, and a dropout additive regression trees classifier with an AUC of 0.8274 and LogLoss of 0.5037. Influential factors included initial pain, disease activity, and Hospital Anxiety and Depression Scale scores for depression and anxiety. Increased pain, disease activity, and psychological distress generally led to higher BASDAI scores. CONCLUSIONS: The practical implications of these models for clinical decision-making in PsV and PsA can guide early investigation and treatment, contributing to improved patient outcomes.

The Effect of an Interdisciplinary Dermatological-Rheumatological Consultation on the Outcome of Patients with Psoriasis with Musculoskeletal Pain: A Prospective, Single-Center Cohort Study.

INTRODUCTION: Psoriatic arthritis (PsA), a disease with complex inflammatory musculoskeletal manifestations, complicates psoriasis in up to 30% of patients. In this study, we aimed to determine the effect of an interdisciplinary dermatological-rheumatological consultation (IDRC) for patients with psoriasis with musculoskeletal symptoms. METHODS: This prospective study enrolled 202 patients with psoriasis. Patients with musculoskeletal pain (MSP) (n = 115) participated in an IDRC 12 weeks after enrollment. The outcome was evaluated after 24 weeks. RESULTS: In 12/79 (15.2%) patients seen in the IDRC, the prior diagnosis was changed: eight with a first diagnosis of PsA, four with a diagnosis of PsA rescinded. Treatment was modified in 28% of patients. Significant improvements in Psoriasis Area and Severity Index (PASI) (from 5.3 to 2.0; p < 0.001) and Dermatology Life Quality Index (DLQI) (from 6.7 to 4.5; p = 0.009) were observed. By comparing changes in PASI and DLQI over the study period, an improvement in PASI of 0.7 ± 1.4 points (p = 0.64) and in DLQI of 2.9 ± 1.5 points (p = 0.051) could be attributed to participation in the IDRC. CONCLUSION: An IDRC of patients with psoriasis with MSP leads to a valid diagnosis of PsA and improvement in quality of life. Based on these results, an IDRC is a valuable and time efficient way for psoriasis patient with MSP to receive optimal care.

German Mobile Apps for Patients With Psoriasis: Systematic Search and Evaluation.

BACKGROUND: Psoriasis is a chronic inflammatory skin disease. The visibility of erythematous plaques on the skin as well as the pain and itchiness caused by the skin lesions frequently leads to psychological distress in patients. Smartphone apps are widespread and easily accessible. Earlier studies have shown that apps can effectively complement current management strategies for patients with psoriasis. However, no analysis of such apps has been published to date. OBJECTIVE: The aim of this study is to systematically identify and objectively assess the quality of current publicly available German apps for patients with psoriasis using the Mobile Application Rating Scale (MARS) and compile brief ready-to-use app descriptions. METHODS: We conducted a systematic search and assessment of German apps for patients with psoriasis available in the Google Play Store and Apple App Store. The identified apps were randomly assigned to 1 of 3 reviewers, who independently rated them using the German MARS (MARS-G). The MARS-G includes 15 items from 4 different sections (engagement, functionality, aesthetics, and information) to create an overall mean score for every app. Scores can range from 1 for the lowest-quality apps to 5 for the highest-quality apps. Apps were ranked according to their mean MARS-G rating, and the highest-ranked app was evaluated independently by 2 patients with psoriasis using the user version of the MARS-G (uMARS-G). Furthermore, app information, including origin, main function, and technical aspects, was compiled into a brief overview. RESULTS: In total, we were able to identify 95 unique apps for psoriasis, of which 15 were available in both app stores. Of these apps, 5 were not specifically intended for patients with psoriasis, 1 was designed for clinical trials only, and 1 was no longer available at the time the evaluation process began. Consequently, the remaining 8 apps were included in the final evaluation. The mean MARS-G scores ranged from 3.51 to 4.18. The app with the highest mean MARS-G score was Psoriasis Helferin (4.18/5.00). When rated by patients, however, the app was rated lower in all subcategories, resulting in a mean uMARS-G score of 3.48. Most apps had a commercial background and a focus on symptom tracking. However, only a fraction of the apps assessed used validated instruments to measure the user's disease activity. CONCLUSIONS: App quality was heterogeneous, and only a minority of the identified apps were available in both app stores. When evaluated by patients, app ratings were lower than when evaluated by health care professionals. This discrepancy highlights the importance of involving patients when developing and evaluating health-related apps as the factors that make an app appealing to users may differ between these 2 groups. TRIAL REGISTRATION: Deutsches Register Klinischer Studien DRKS00020963; https://tinyurl.com/ye98an5b.

From lignins to tannins: forty years of enzyme studies on the biosynthesis of phenolic compounds.

In the early 1960s, enzyme studies increasingly began to replace the common 'feeding' experiments in which labeled tracers were applied to living plants or plant parts for elucidating metabolic pathways. This advanced technique allowed to gain much deeper insights into individual details of metabolic sequences, and particularly on the previously inaccessible role of activated 'energy-rich' intermediates. Based on the author's own experience for the past 40+ years in this field, principal findings and trends elucidating the pathways to lignin and lignin precursors, acyl amides and hydrolyzable tannins (gallotannins, ellagitannins) by enzyme studies are reported.

Enzymology of gallotannin and ellagitannin biosynthesis.

Gallotannins and ellagitannins, the two subclasses of hydrolyzable tannins, are derivatives of 1,2,3,4,6-penta-O-galloyl-beta-D-glucopyranose. Enzyme studies with extracts from oak leaves (Quercus robur, syn. Quercus pedunculata; Quercus rubra) and from staghorn sumac (Rhus typhina) revealed that this pivotal intermediate is synthesized from beta-glucogallin (1-O-galloyl-beta-D-glucopyranose) by a series of strictly position-specific galloylation steps, affording so-called 'simple' gallotannins, i.e., mono- to pentagallyoylglucose esters. Besides its role as starter molecule, beta-glucogallin was also recognized as the principal energy-rich acyl donor required in these transformations. Subsequent pathways to 'complex' gallotannins have recently been elucidated by the isolation of five different enzymes from sumac leaves that were purified to apparent homogeneity. They catalyzed the beta-glucogallin-dependent galloylation of pentagallyoylglucose to a variety of hexa- and heptagalloylglucoses, plus several not yet characterized higher substituted analogous galloylglucoses. With respect to the biosynthesis of ellagitannins, postulates that had been formulated already decades ago were proven by the purification of a new laccase-like phenol oxidase from leaves of fringe cups (Tellima grandiflora) that regio- and stereospecifically oxidized pentagallyoylglucose to the monomeric ellagitannin, tellimagrandin II. This compound was further oxidized by a similar but different laccase-like oxidase to yield a dimeric ellagitannin, cornusiin E.

Oxidation of pentagalloylglucose to the ellagitannin, tellimagrandin II, by a phenol oxidase from Tellima grandiflora leaves.

A new enzyme has been isolated from leaves of the weed Tellima grandiflora (fringe cups, Saxifragaceae) that catalyzed the O(2)-dependent oxidation of 1,2,3,4,6-penta-O-galloyl-beta-D-glucopyranose to tellimagrandin II, the first intermediate in the (4)C(1)-glucose derived series of ellagitannins. CD-spectra revealed that the 4,6-O-HHDP-residue of the in vitro product had the (S)-stereoconfiguration characteristic of tellimagrandin II from natural sources. The enzyme, for which a M(r) of ca. 60,000 was determined, was purified to apparent homogeneity. It had a pH-optimum at pH 5.0, an isoelectric point at pH 6.3 and was most stable at pH 4.2. Inhibition studies suggested that this new enzyme, for which the systematic name 'pentagalloylglucose: O(2) oxidoreductase' is proposed, belongs to the vast group of laccase-type phenol oxidases (EC 1.10.3.2).

Ellagitannin biosynthesis: laccase-catalyzed dimerization of tellimagrandin II to cornusiin E in Tellima grandiflora.

An enzyme has been purified from leaves of the weed Tellima grandiflora (fringe cups, Saxifragaceae) that catalyzed the O2-dependent oxidation of the monomeric ellagitannin, tellimagrandin II, to a dimeric derivative, cornusiin E. The apparently homogeneous enzyme preparation had a Mr of ca. 160,000 (with four subunits of Mr 40,000), a pH-optimum and an isoelectric point at pH 5.2, and was most stable at pH 4.3. Inhibition studies revealed that this new enzyme, for which the systematic name 'tellimagrandin II: O2 oxidoreductase' is proposed, is a member of the laccase (EC 1.10.3.2) family of phenol oxidases. The properties of this enzyme differed from that of a related laccase that catalyzed the transition of 1,2,3,4,6-pentagalloylglucopyranose to tellimagrandin II, the preceding step in the biosynthetic route to cornusin E.

Biosynthesis of the dimeric ellagitannin, cornusiin E, in Tellima grandiflora.

First evidence for the in vitro synthesis of a dimeric ellagitannin has been obtained with cell-free extracts from the weed Tellima grandiflora (fringe cups, Saxifragaceae). Partially purified enzyme preparations from leaves of this plant catalyzed the oxidation of 1,2,3,4,6-pentagalloyl-beta-D-glucose to the monomeric ellagitannin, tellimagrandin II, followed by oxidative coupling of two units of this intermediate to yield a dimeric derivative. Chemical degradation, MALDI-TOF mass spectrometry, 1H and 13C nuclear magnetic resonance, and CD spectroscopy were employed to identify this enzyme reaction product as cornusiin E which is characterized by a (S)-valoneoyl bridge between glucose-positions 2, 4' and 6'. This result was supported by comparison with data obtained for cornusiin E that had been isolated from leaves of intact T. grandiflora plants. No indication for the earlier proposed existence of rugosin D (an isomer with a 1,4',6'-bound valoneoyl unit) in T. grandiflora has been obtained in this investigation.

Gallotannin biosynthesis: two new galloyltransferases from Rhus typhina leaves preferentially acylating hexa- and heptagalloylglucoses.

Current enzyme studies on the biosynthesis of gallotannins with cell-free extracts from leaves of staghorn sumac (Rhus typhina L.) revealed the existence of two new beta-glucogallin-dependent galloyltransferases (EC 2.3.1.-) that preferentially catalyzed the acylation of hexa- and heptagalloylglucoses. One enzyme was most active with the hexagalloylglucose, 3-O-digalloyl-1,2,4,6-tetra-O-galloylglucose, to form the corresponding heptagalloylglucose, 3-O-trigalloyl-1,2,4,6-tetra-O-galloylglucose. This polyester, in turn, was the preferred substrate for a second enzyme that catalyzed its conversion to higher substituted derivatives. This latter enzyme also displayed considerable affinity towards 2-O-digalloyl-1,3,4,6-tetra-O-galloylglucose which was acylated to various hepta- and octagalloylglucoses. These recent findings, together with data from earlier reported related enzymes, allowed the presentation of a scheme that summarizes the major transitions in the biogenetic routes from 1,2,3,4,6-pentagalloylglucose to complex gallotannins.