Artificial intelligence for predictive biomarker discovery in immuno-oncology: a systematic review

BACKGROUND

The widespread use of immune checkpoint inhibitors (ICIs) has revolutionised treatment of multiple cancer types. However, selecting patients who may benefit from ICI remains challenging. Artificial intelligence (AI) approaches allow exploitation of high-dimension oncological data in research and development of precision immuno-oncology.

MATERIALS AND METHODS

We conducted a systematic literature review of peer-reviewed original articles studying the ICI efficacy prediction in cancer patients across five data modalities: genomics (including genomics, transcriptomics, and epigenomics), radiomics, digital pathology (pathomics), and real-world and multimodality data.

RESULTS

A total of 90 studies were included in this systematic review, with 80% published in 2021-2022. Among them, 37 studies included genomic, 20 radiomic, 8 pathomic, 20 real-world, and 5 multimodal data. Standard machine learning (ML) methods were used in 72% of studies, deep learning (DL) methods in 22%, and both in 6%. The most frequently studied cancer type was non-small-cell lung cancer (36%), followed by melanoma (16%), while 25% included pan-cancer studies. No prospective study design incorporated AI-based methodologies from the outset; rather, all implemented AI as a post hoc analysis. Novel biomarkers for ICI in radiomics and pathomics were identified using AI approaches, and molecular biomarkers have expanded past genomics into transcriptomics and epigenomics. Finally, complex algorithms and new types of AI-based markers, such as meta-biomarkers, are emerging by integrating multimodal/multi-omics data.

CONCLUSION

AI-based methods have expanded the horizon for biomarker discovery, demonstrating the power of integrating multimodal data from existing datasets to discover new meta-biomarkers. While most of the included studies showed promise for AI-based prediction of benefit from immunotherapy, none provided high-level evidence for immediate practice change. A priori planned prospective trial designs are needed to cover all lifecycle steps of these software biomarkers, from development and validation to integration into clinical practice.

Upper limb soft robotic wearable devices: a systematic review

INTRODUCTION

Soft robotic wearable devices, referred to as exosuits, can be a valid alternative to rigid exoskeletons when it comes to daily upper limb support. Indeed, their inherent flexibility improves comfort, usability, and portability while not constraining the user's natural degrees of freedom. This review is meant to guide the reader in understanding the current approaches across all design and production steps that might be exploited when developing an upper limb robotic exosuit.

METHODS

The literature research regarding such devices was conducted in PubMed, Scopus, and Web of Science. The investigated features are the intended scenario, type of actuation, supported degrees of freedom, low-level control, high-level control with a focus on intention detection, technology readiness level, and type of experiments conducted to evaluate the device.

RESULTS

A total of 105 articles were collected, describing 69 different devices. Devices were grouped according to their actuation type. More than 80% of devices are meant either for rehabilitation, assistance, or both. The most exploited actuation types are pneumatic (52%) and DC motors with cable transmission (29%). Most devices actuate 1 (56%) or 2 (28%) degrees of freedom, and the most targeted joints are the elbow and the shoulder. Intention detection strategies are implemented in 33% of the suits and include the use of switches and buttons, IMUs, stretch and bending sensors, EMG and EEG measurements. Most devices (75%) score a technology readiness level of 4 or 5.

CONCLUSION

Although few devices can be considered ready to reach the market, exosuits show very high potential for the assistance of daily activities. Clinical trials exploiting shared evaluation metrics are needed to assess the effectiveness of upper limb exosuits on target users.

Increased task-uncorrelated muscle activity in childhood dystonia

BACKGROUND

Even if movement abnormalities in dystonia are obvious on observation-based examinations, objective measures to characterize dystonia and to gain insights into its pathophysiology are still strongly needed. We hypothesize that motor abnormalities in childhood dystonia are partially due to the inability to suppress involuntary variable muscle activity irrelevant to the achievement of the desired motor task, resulting in the superposition of unwanted motion components on the desired movement. However, it is difficult to separate and quantify appropriate and inappropriate motor signals combined in the same muscle, especially during movement.

METHODS

We devise an innovative and practical method to objectively measure movement abnormalities during the performance of a continuous figure-eight writing task in 7 children with dystonia and 9 age-matched healthy controls. During the execution of a continuous writing task, muscle contractions should occur at frequencies that match the frequencies of the writing outcome. We compare the power spectra of kinematic trajectories and electromyographic signals of 8 upper limb muscles to separate muscle activity with the same frequency content of the figure-eight movement (task-correlated) from activity occurring at frequencies extraneous to the task (task-uncorrelated).

RESULTS

Children with dystonia present a greater magnitude of task-uncorrelated muscle components. The motor performance achieved by children with dystonia is characterized by an overall lower quality, with high spatial and temporal variability and an altered trade-off between speed and accuracy.

CONCLUSIONS

Findings are consistent with the hypothesis that, in childhood dystonia, the ability to appropriately suppress variable and uncorrelated elements of movement is impaired. Here we present a proof-of-concept of a promising tool to characterize the phenomenology of movement disorders and to inform the design of neurorehabilitation therapies.