*Article* **A Novel Digital Care Management Platform to Monitor Clinical and Subclinical Disease Activity in Multiple Sclerosis**

**Wim Van Hecke 1,2,\* ,†, Lars Costers 1,2,† , Annabel Descamps <sup>1</sup> , Annemie Ribbens <sup>1</sup> , Guy Nagels 1,2,3 , Dirk Smeets 1,2 and Diana M. Sima 1,2**


**Abstract:** In multiple sclerosis (MS), the early detection of disease activity or progression is key to inform treatment changes and could be supported by digital tools. We present a novel CEmarked and FDA-cleared digital care management platform consisting of (1) a patient phone/web application and healthcare professional portal (ico**mpanion**) including validated symptom, disability, cognition, and fatigue patient-reported outcomes; and (2) clinical brain magnetic resonance imaging (MRI) quantifications (ico**brain ms**). We validate both tools using their ability to detect (sub)clinical disease activity (known-groups validity) and real-world data insights. Surveys showed that 95.6% of people with MS (PwMS) were interested in using an MS app, and 98.2% were interested in knowing about MRI changes. The ico**mpanion** measures of disability (*p* < 0.001) and symptoms (*p* = 0.005) and ico**brain ms** MRI parameters were sensitive to (sub)clinical differences between MS subtypes. ico**brain ms** also decreased intra- and inter-rater lesion count variability and increased sensitivity for detecting disease activity/progression from 24% to 76% compared to standard radiological reading. This evidence shows PwMS' interest, the digital care platform's potential to improve the detection of (sub)clinical disease activity and care management, and the feasibility of linking different digital tools into one overarching MS care pathway.

**Keywords:** multiple sclerosis; ico**mpanion**; ico**brain**; eHealth; digital health technology; mobile application; patient reported outcomes; magnetic resonance imaging

#### **1. Introduction**

Today, more than 2.8 million people are living with multiple sclerosis (MS), making it the most common progressive neurological condition in young people [1]. MS is characterized either by periods of relapses and remission or a progressive disability pattern. Currently, there are over 20 disease-modifying treatments (DMTs) available, aiming to slow down relapses and disease progression [2]. Thanks to these DMTs, the health of people with MS (PwMS), expressed in quality-adjusted life years (QALYs), has been estimated to have increased by 66% since the launch of the first drug in 1993 [3]. However, despite the increased availability of DMTs, 26% to 40% of PwMS are estimated to be on a suboptimal treatment [4,5].

These findings illustrate the challenge in MS care, which is providing individual PwMS with the right drug at the right time. Hence, in order to make informed treatment decisions, it is crucial to measure disease activity and progression in a standardized manner. In this context, disease activity and progression are typically evaluated by the clinical assessment of relapses and disability (measured by the Expanded Disability Status Scale (EDSS)), and

**Citation:** Van Hecke, W.; Costers, L.; Descamps, A.; Ribbens, A.; Nagels, G.; Smeets, D.; Sima, D.M. A Novel Digital Care Management Platform to Monitor Clinical and Subclinical Disease Activity in Multiple Sclerosis. *Brain Sci.* **2021**, *11*, 1171. https://doi.org/10.3390/ brainsci11091171

Academic Editors: Tjalf Ziemssen and Rocco Haase

Received: 31 July 2021 Accepted: 1 September 2021 Published: 3 September 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

longitudinal changes on the brain magnetic resonance imaging (MRI) scans (looking at new/enlarging lesions and/or brain atrophy) [6].

However, it is known that clinical disease activity and progression often go unnoticed during clinical assessment, partly due to the problem that relapses are systematically underreported by around half of PwMS [6,7]. In addition, MS progression goes beyond relapses and physical disability worsening, as problems with memory but also linguistic and verbal fluency problems are known to be important components of MS-related disability [8,9]. These components are often not routinely assessed during the patient visit or are based on the PwMS' recollection on how they have been doing since the last visit. In addition, it has been demonstrated that there is a significant clinician-dependent variability in the assessment of MS patient disease activity [10].

The second component of evaluating disease activity is based on the assessment of subclinical progression on brain MRI scans. International guidelines recommend the acquisition of brain (and increasingly spinal cord) MRI scans for diagnosis and a yearly scan for follow-up [11]. However, in a clinical setting, brain MRI reading is known to be qualitative, based on a visual assessment, and radiologist-dependent, leading to discrepancies in the radiological reports [12]. Indeed, it has been reported that up to 24% of brain MRI reporting contains discrepancies when reviewed by a panel of radiologists [12].

There is great promise in implementing digital health solutions to standardize MS care, as they can improve efficiency and workflow, and complement clinical expertise. A recent study indicated healthcare professionals' (HCPs) most crucial problems in MS disease management to be the lack of forwarding of information by the patient, the need for the patient to visit on site for inquiries and poor reachability of PwMS, for which digital telemonitoring tools can be a solution [13].

Remote patient monitoring through medical health (mHealth) applications in MS care allows for a more continuous and data-driven monitoring of symptoms and disease progression [14,15]. Regular standardized check-ins through mHealth apps have the potential to mitigate the underreporting of important clinical events [7] and bridge the information gap between annual neurology visits. It has been estimated that personalized medicine tools in MS have the potential to increase the impact of treatments by more than 50% by quantifying both disease activity (clinical and subclinical) and the risk of side effects [3]. In addition to the value of mHealth tools for health-care professionals (HCPs), there is also the potential to further empower PwMS, resulting in an increased self-management and allowing more open and early conversations about disease progression [16,17].

In addition to mHealth applications, artificial intelligence (AI) solutions have been developed to detect and quantify disease activity on MRI scans, which play a central role in disease management. In the last decades, several software tools have been developed and applied for research and clinical trials. Examples of widely used neuroimage analysis packages for research purposes include Freesurfer (https://surfer.nmr.mgh.harvard.edu accessed on 27 August 2021), FMRIB Software Library (FSL; https//fsl.fmrib.ox.ac.uk/fsl, accessed on 27 August 2021), and Statistical Parametric Mapping (SPM; https://www. fil.ion.ucl.ac.uk/spm, accessed on 27 August 2021). However, only very few brain MRI solutions exist that have been thoroughly validated and cleared as a medical device for clinical use [18].

In this paper, we present a novel digital care management platform for MS that aims at standardizing MS patient care and allowing more data-driven clinical decisions in the MS care pathway. The platform includes a CE marked and FDA cleared mHealth application that collects patient-reported information in the period of time between neurology visits, a CE marked and FDA cleared solution that quantifies clinically relevant brain MRI changes in PwMS, and the necessary software solutions that guarantee a seamless integration of the digital MS care management platform into the clinical workflow. We investigate the needs and interests of PwMS concerning such solutions, and their potential to improve the detection of (sub-)clinical disease activity and care management of MS.

#### **2. Materials and Methods** an FDA class 2 medical device and MDD class 1m medical device. ico**metrix**' secure cloud

**2. Materials and Methods** 

*2.1. MS Care Management Platform* 

#### *2.1. MS Care Management Platform* is ISO13485 and ISO27001 certified and GDPR and HIPAA compliant regarding the Secu-

As illustrated in Figure 1, the care management platform consists of multiple components: (1) the ico**mpanion** patient mobile phone application (available on Android and iOS) and website (accessible via web browser: icompanion.ms), (2) the ico**mpanion** web portal for HCPs (accessible via web browser), (3) the ico**brain ms** volumetric brain reports and (4) integrations with hospitals' Picture Archiving and Communication System (PACS) and electronic medical record (EMR) systems. rity and Privacy Rules. Incoming DICOM files of MRI scans are pseudonymized according to HIPAA standard and all fields containing private patient information are removed, except patient gender and birth date (transformed to YYYY-01-01) in order to be able to provide a correct analysis and compare the patient with a healthy population.

As illustrated in Figure 1, the care management platform consists of multiple components: (1) the ico**mpanion** patient mobile phone application (available on Android and iOS) and website (accessible via web browser: icompanion.ms), (2) the ico**mpanion** web portal for HCPs (accessible via web browser), (3) the ico**brain ms** volumetric brain reports and (4) integrations with hospitals' Picture Archiving and Communication System

Both ico**mpanion** and ico**brain ms** are registered medical devices and were developed by ico**metrix** (Leuven, Belgium). According to FDA regulation, ico**mpanion** is a class 1 medical device, and under EU MDD regulation a class 1 medical device. ico**brain ms** is

*Brain Sci.* **2021**, *11*, x FOR PEER REVIEW 3 of 27

(PACS) and electronic medical record (EMR) systems.

**Figure 1.** The MS care management platform consists of the ico**mpanion** patient app and website, ico**mpanion** HCP web **Figure 1.** The MS care management platform consists of the ico**mpanion** patient app and website, ico**mpanion** HCP web portal, integration with ico**brain ms** volumetric brain reports and integration with hospital's electronic medical records.

portal, integration with ico**brain ms** volumetric brain reports and integration with hospital's electronic medical records. 2.1.1. icom**panion** Patient App and Website Using the ico**mpanion** app and website, PwMS can keep a diary, log symptoms, and perform tests for body function, cognitive function, and fatigue (Figure 2) based on clinically validated patient reported outcomes (PROs) described below [19–22], which can be shared with the patient's clinical team. In addition, PwMS can add treatment information, from DMTs to symptomatic and rehabilitation treatments, and set reminders on when to Both ico**mpanion** and ico**brain ms** are registered medical devices and were developed by ico**metrix** (Leuven, Belgium). According to FDA regulation, ico**mpanion** is a class 1 medical device, and under EU MDD regulation a class 1 medical device. ico**brain ms** is an FDA class 2 medical device and MDD class 1m medical device. ico**metrix**' secure cloud is ISO13485 and ISO27001 certified and GDPR and HIPAA compliant regarding the Security and Privacy Rules. Incoming DICOM files of MRI scans are pseudonymized according to HIPAA standard and all fields containing private patient information are removed, except patient gender and birth date (transformed to YYYY-01-01) in order to be able to provide a correct analysis and compare the patient with a healthy population.

#### take or perform their treatment. Furthermore, PwMS can easily upload their MRIs (via the 2.1.1. icom**panion** Patient App and Website

patient website) and view them (via patient website and app) as well as learn about topics related to MS (e.g., MS types, MRI, lesions). Finally, PwMS can prepare their consultations using a pre-visit checklist, the answers of which are also shared with the patient's clinical team (e.g., 'Do you need any new prescription, certificates or reimbursement documents?'). The clinically validated PROs included into ico**mpanion** are the SymptoMScreen, a patient-reported Expanded Disability Status Scale (prEDSS), Neuro-QoL (V1.0) Fatigue short-form, and the Neuro-QoL (V2.0) Cognitive Function short-form: Using the ico**mpanion** app and website, PwMS can keep a diary, log symptoms, and perform tests for body function, cognitive function, and fatigue (Figure 2) based on clinically validated patient reported outcomes (PROs) described below [19–22], which can be shared with the patient's clinical team. In addition, PwMS can add treatment information, from DMTs to symptomatic and rehabilitation treatments, and set reminders on when to take or perform their treatment. Furthermore, PwMS can easily upload their MRIs (via the patient website) and view them (via patient website and app) as well as learn about topics related to MS (e.g., MS types, MRI, lesions). Finally, PwMS can prepare their consultations using a pre-visit checklist, the answers of which are also shared with the patient's clinical team (e.g., 'Do you need any new prescription, certificates or reimbursement documents?').

• The SymptoMScreen [21] is a 12-item battery for MS-related symptoms with a 7-point

Likert scale per functional domain. Scores range from 'not affected at all' (score = 0) to 'total limitation' (score = 6). The SymptoMScreen composite is a score that

**Figure 2.** Overview of the ico**mpanion** patient app (**a**–**g**) and HCP platform features (**h**,**i**): (**a**) easy check-in and diary, (**b**) PROs for symptoms, EDSS, fatigue, cognition, . . . , (**c**) treatment logging and reminders, (**d**) preparation neurologist visit, (**e**) knowledge center, (**f**) MRI viewer, (**g**) linking with MS team, (**h**) interactive overview of PRO data and downloadable reports and (**i**) automatic import of MRI scans from hospital PACS system and integration with ico**brain ms** reports.

1

The clinically validated PROs included into ico**mpanion** are the SymptoMScreen, a patient-reported Expanded Disability Status Scale (prEDSS), Neuro-QoL (V1.0) Fatigue short-form, and the Neuro-QoL (V2.0) Cognitive Function short-form:


### 2.1.2. ico**brain ms** Volumetric MRI Analyses

ico**brain ms** is an AI software solution for brain magnetic resonance image analysis in MS, producing annotated images and pre-populated radiological reports (icometrix.com/services/icobrainms accessed on 27 August 2021). The main components of ico**brain ms** are the brain tissue segmentation and MS lesion segmentation on single-time point T1-weighted and FLAIR scans, as well as specific longitudinal volume change computations for establishing brain atrophy rates and lesion evolution. Whole brain and gray matter volumes, normalized for head size, are compared against age- and sex-matched reference controls populations. In Figure 3, an example of the ico**brain ms** output is shown, including the annotated images, the quantitative reports, and the pre-populated structured radiological report that are provided in the local PACS system and available by the time the radiological reading starts. In the top row of Figure 3, a sagittal slice of the annotated images (which are presented in the same space as the original scans) is shown, the bottom row includes the quantitative ico**brain ms** reports and an example of the pre-populated radiological report:


The technical details, validation, and clinical usefulness of the methodology have been published previously [24–28]. The software is seamlessly integrated in the clinical workflow via ico**bridge** (see Section 2.1.3 or https://icobridge.icometrix.com accessed on 27 August 2021). It includes direct and secure upload from the hospital's image archiving system to ico**metrix**' servers, where the AI pipeline is run, and secure transfer of reports and annotated images back to the hospital's system in time for the radiological reading.

**Figure 3.** Output of the ico**brain ms** software, which is integrated with the local PACS. In the top row, a sagittal slice of the annotated images (which are presented in the same space as the original **Figure 3.** Output of the ico**brain ms** software, which is integrated with the local PACS. In the top row, a sagittal slice of the annotated images (which are presented in the same space as the original scans) is shown. The bottom row includes the quantitative ico**brain ms** reports and an example of the pre-populated radiological report.

#### 2.1.3. ico**mpanion** HCP Portal

2.1.2. ico**brain ms** Volumetric MRI Analyses

populated radiological report:

fratentorial (green);

sions and their location.

guage.

ico**brain ms** is an AI software solution for brain magnetic resonance image analysis in MS, producing annotated images and pre-populated radiological reports (icometrix.com/services/icobrain-ms accessed on 27 August 2021). The main components of ico**brain ms** are the brain tissue segmentation and MS lesion segmentation on single-time point T1-weighted and FLAIR scans, as well as specific longitudinal volume change computations for establishing brain atrophy rates and lesion evolution. Whole brain and gray matter volumes, normalized for head size, are compared against age- and sex-matched reference controls populations. In Figure 3, an example of the ico**brain ms** output is shown, including the annotated images, the quantitative reports, and the pre-populated structured radiological report that are provided in the local PACS system and available by the time the radiological reading starts. In the top row of Figure 3, a sagittal slice of the annotated images (which are presented in the same space as the original scans) is shown, the bottom row includes the quantitative ico**brain ms** reports and an example of the pre-

• top left = T1 overlaid with gray matter segmentation (blue) and T1 lesions (red); • top middle = FLAIR overlaid with lesion segmentations color coded by location: periventricular (yellow), deep white matter (blue), juxtacortical (purple), and in-

• top right = FLAIR overlaid with color coded lesion changes compared to last available (or selected) scan: existing (green), enlarging (orange), new (red) lesions. • bottom left = the quantitative report of whole brain and gray matter volume and at-

• bottom middle = the quantitative report of existing, enlarging and, new FLAIR le-

which already includes the ico**brain ms** measures and is available in the local lan-

rophy (and comparison with healthy population).

In the HCP portal, HCPs can access the entered ico**mpanion** PRO data from their linked PwMS as well as their MRI images and ico**brain ms** volumetric brain reports (Figures 1–3). Access to the data on this portal can be easily managed via the MS team functionality, which allows the set-up of a care team with different team members and roles. The entered ico**mpanion** PRO data can be viewed in an interactive plot with tools to help the interpretation, and HCPs can download pdf reports. HCPs can also view the PwMS' uploaded MRIs, and using ico**bridge** (https://icobridge.icometrix.com accessed on 27 August 2021), ico**metrix**' secure DICOM gateway, MRI scans can also be automatically imported from a hospital's PACS system. All imported MRI scans are analyzed using the ico**brain ms** volumetric analysis, after which a report can be downloaded from the HCP portal (Figure 3). The ico**mpanion** and ico**brain ms** reports (and raw data points or intermediate results) can be imported into a hospital's EMR system. Using the HL7 v2 protocol, these reports can be sent over to the EMR, either in their native pdf format, coded values or as simple text. From there on this data can for instance be shown in a radiologic report or attached to a study for future reference.

#### *2.2. Patient Perspective*

#### 2.2.1. Patient Survey 1: Telemonitoring Tools for Monitoring Clinical Disease Activity

In order to develop a patient monitoring tool that responds to the needs of PwMS and fits PwMS' everyday life, a survey was sent out to PwMS with MS via local patient support groups without any randomization. In this survey, answered by 45 PwMS, the PwMS were asked about their knowledge about important topics such as MRI, EDSS, etc., which was used to develop educational content for ico**mpanion**. Next, the PwMS were asked about their attitude towards an app to monitor MS, different possible features, and their interest in using such an app.

#### 2.2.2. Patient Survey 2: MR Imaging for Monitoring Subclinical Disease Activity

Together with iConquerMS (iConquerMS.org), a survey was sent out in June 2020 [29], with questions about MRI access, viewing and knowledge, which was answered by 876 PwMS. As an example, questions included (see Appendices A and B for the complete list of questions and possible answers in the survey):

