Augmented Versus Artificial Intelligence for Stratification of Patients with Myositis

With interest we read the recent article by Pinal-Fernandez and Mammen,1 which comments on the paper by Spielmann et al2 and to a lesser extent on the contribution by Mariampillai et al3 4 and raises concerns about the artificial intelligence (AI)-driven approach used to define subgroups of patients with idiopathic inflammatory myopathy (IIM).

To illustrate this, Pinal-Fernandez and Mammen constructed a library of 1000 observations and selected the four variables using a multivariate normal distribution, thus finding a similar clustering as in the original paper by Spielmann et al.2 We share some of the concerns about unsupervised learning techniques raised by Pinal-Fernandez and Mammen.1 In this letter, we would like to highlight several aspects related to AI-driven methodologies.

Machine learning (ML) is a subset of AI that enables a computer to make decisions based on the large dataset. When applied to clustering, it will always give an ‘optimal’ solution for the number of clusters ‘present’ in a dataset. However, it is up to the human user’s discretion to determine whether those clusters exist. An ML algorithm determines a number of clusters by separating the datasets into the subgroups through a process of optimising (1) separation between each cluster to its greatest and (2) ensuring that within a cluster, the distance to the cluster centre for each point is the smallest. Such an algorithm is essentially trying to identify a number of optimal clusters that allow each cluster to be distinct from the others. The goal is to have tight individual clusters that are very distinguishable from the others. In any dataset, the algorithms will present an optimal solution to those or similar criteria, but it does not always mean those clusters are truly significant or meaningful.

Visualising the clusters using dimensionality reduction techniques such as principal component analysis or t-distributed stochastic neighbour embedding is vital for this process, in addition to more quantitative methods such as comparing intracluster variation, intercluster variation and silhouette scoring. That is why researchers using ML should ideally be ‘bilingual’ and understand both the mathematics and algorithms, as well the science and clinical meaning behind the results.

To conclude, we emphasise that, no doubt, ML has the potential to improve the stratification of patients with IIM if certain concepts of data science are followed as also pointed out by a task force of the European League Against Rheumatism for big data and AI.5 ML relies on large, standardised and curated datasets that require large patient cohorts. Due to the rarity of IIM, larger patient cohorts (such as the MyoNet/EuroMyositis)6 are required to generate quality data. Once larger and curated datasets are available, the ML approach is a powerful alternative to human judgement and can improve future classification criteria for IIM.4 7 8Today, we argue for the use of ML alongside expert decision, thus relying on augmented judgement when making the final decision on patient stratification especially when building AI-based models. Augmented intelligence has the potential for improved patient stratification in IIM.

A Novel Amino Acid Composition Ameliorates Short-Term Muscle Disuse Atrophy in Healthy Young Men

Skeletal muscle disuse leads to atrophy, declines in muscle function, and metabolic dysfunction that are often slow to recover. Strategies to mitigate these effects would be clinically relevant. In a double-blind randomized-controlled pilot trial, we examined the safety and tolerability as well as the atrophy mitigating effect of a novel amino acid composition (AXA2678), during single limb immobilization. Twenty healthy young men were randomly assigned (10 per group) to receive AXA2678 or an excipient- and energy-matched non-amino acid containing placebo (PL) for 28d: days 1–7, pre-immobilization; days 8–15, immobilization; and days 16–28 post-immobilization recovery. Muscle biopsies were taken on d1, d8 (immobilization start), d15 (immobilization end), and d28 (post-immobilization recovery). Magnetic resonance imaging (MRI) was utilized to assess quadriceps muscle volume (Mvol), muscle cross-sectional area (CSA), and muscle fat-fraction (FF: the fraction of muscle occupied by fat). Maximal voluntary leg isometric torque was assessed by dynamometry. Administration of AXA2678 attenuated muscle disuse atrophy compared to PL (p < 0.05) with changes from d8 to d15 in PL: ΔMvol = −2.4 ± 2.3% and ΔCSA = −3.1% ± 2.1%, both p < 0.001 vs. zero; against AXA2678: ΔMvol: −0.7 ± 1.8% and ΔCSA: −0.7 ± 2.1%, both p > 0.3 vs. zero; and p < 0.05 between treatment conditions for CSA. During immobilization, muscle FF increased in PL but not in AXA2678 (PL: 12.8 ± 6.1%, AXA2678: 0.4 ± 3.1%; p < 0.05). Immobilization resulted in similar reductions in peak leg isometric torque and change in time-to-peak (TTP) torque in both groups. Recovery (d15–d28) of peak torque and TTP torque was also not different between groups, but showed a trend for better recovery in the AXA2678 group. Thrice daily consumption of AXA2678 for 28d was found to be safe and well-tolerated. Additionally, AXA2678 attenuated atrophy, and attenuated accumulation of fat during short-term disuse. Further investigations on the administration of AXA2678 in conditions of muscle disuse are warranted.

Towards Objective and Reproducible Measures of Thigh Muscle Fat Fraction in Patients with Duchenne Muscular Dystrophy

Towards Objective and Reproducible Measures of Thigh Muscle Fat Fraction in Patients with Duchenne Muscular Dystrophy

Maecenas tempor, ex at maximus efficitur, felis sem ultrices ligula, ut hendrerit purus eros ac urna. Vestibulum ut vestibulum tortor. Orci varius natoque penatibus et magnis dis parturient montes, nascetur ridiculus mus. Cras eu lectus quam. Nunc ligula arcu, auctor sit amet tellus eleifend, facilisis laoreet odio. Donec placerat urna eleifend blandit porttitor.

Nunc magna turpis, tristique at dictum vel, sollicitudin blandit felis. Morbi aliquam elit et pellentesque vulputate. Donec elementum, ante quis ornare porttitor, tortor dolor vestibulum velit, et viverra enim massa vitae ex.