=Paper=
{{Paper
|id=Vol-3762/478
|storemode=property
|title=ARTIS: a digital interface to promote the rehabiliatation of text comprehension difficulties through Artificial Intelligence
|pdfUrl=https://ceur-ws.org/Vol-3762/478.pdf
|volume=Vol-3762
|authors=Martina Galletti,Eleonora Pasqua,Manuela Calanca,Caterina Marchesi,Donatella Tomaiuoli,Daniele Nardi
|dblpUrl=https://dblp.org/rec/conf/ital-ia/AgateCPFGRM24a
}}
==ARTIS: a digital interface to promote the rehabiliatation of text comprehension difficulties through Artificial Intelligence==
https://ceur-ws.org/Vol-3762/478.pdf
ARTIS: a digital interface to promote the rehabiliatation of
text comprehension difficulties through Artificial
Intelligence
Martina Galletti1,2,*,† , Eleonora Pasqua3,2,† , Manuela Calanca3,2 , Caterina Marchesi3 ,
Donatella Tomaiuoli3,2,5 and Daniele Nardi2,4
1
Sony Computer Science Laboratories-Paris (Sony CSL-Paris) - France
2
Sapienza University of Rome - Italy
3
Centro Ricerca e Cura di Roma - Italy
5
Tor Vergata University of Rome, Italy
4
CINI-AIIS - Italy
Abstract
This study introduces ARTIS, an AI-powered interface aimed at enhancing text comprehension in children with language and
learning disorders and deficits. Leveraging neuro-psycholinguistic models, ARTIS offers personalized practice at varying
difficulty levels, targeting specific linguistic components of text processing. Through AI algorithms, ARTIS autonomously
extracts keywords, associates them with pictograms, identifies complex words, generates semantic networks, and proposes
exercises on grammatical components. By automating cognitive processes and providing tailored interventions, ARTIS
represents a significant advancement in promoting inclusive hybrid speech and language therapy practices, and improving
text comprehension skills in children with special educational needs.
Keywords
Natural Language Processing, Text comprehension, Language and learning disorders, Speech and Language Therapy
1. Introduction semantic processing. Finally, [5] and [6] addressed the
issue of grammar, claiming that children and adolescents
Understanding a text allows us to acquire information, with problems in text comprehension show difficulties
access new content, enrich the knowledge we already in understanding the role of pronouns within sentences,
possess. Conversely, having difficulty in understanding especially if these are in clitic form.
what one reads is a great limitation to personal growth, Artificial intelligence-powered clinical and education
and restrict opportunities, not only at school, but also in tools have the potential to revolutionise learning and
life. Text comprehension is thus a fascinating but also speech and language therapy for children with special
very complex ability. It requires motivation, attention, educational needs and disabilities. It can personalize
memory, but also specific language skills. These can be interventions and adaptive content that meet individ-
problematic for children with language and learning dis- ual needs, thus promoting a more inclusive and equi-
orders. In fact, subjects with poor text comprehension table educational experience. Within this context we
show difficulties related to the processing of syntactic developed ARTIS — an interface designed to facilitate
and semantic sentence components [1], the analysis of the rehabilitation and instruction of text comprehension
lexical components of words [2] and deficits in the syn- skills through artificial intelligence. Starting from neuro-
tactic representation of words and oral comprehension psycholinguistic models of reading comprehension and
skills [3]. Moreover, [4] stated that the same subjects focusing on the linguistic components of text processing,
report significant deficits in receptive vocabulary and ARTIS enables personalized practice on texts at different
levels. Thanks to AI algorithms, the interface is able to
Ital-IA 2024: 4th National Conference on Artificial Intelligence, orga-
nized by CINI, May 29-30, 2024, Naples, Italy automatically extract pictograms from keywords, iden-
*
Corresponding author. tify more complex words, generate semantic networks,
†
These authors contributed equally. and to propose exercises on certain grammatical compo-
$ martina.galletti@sony.com (M. Galletti); nents. ARTIS is aimed at primary and secondary school
e.pasqua@crc-balbuzie.it (E. Pasqua); m.calanca@crc-balbuzie.it children with difficulties in understanding text, but can
(M. Calanca); c.marchesi@crc-balbuzie.it (C. Marchesi);
also be used as support for English as an L2. The output
d.tomaiuoli@crc-balbuzie.it (D. Tomaiuoli); nard@diag.uniroma1.it
(D. Nardi) of our contribution is twofold. First, we designed and
0009-0002-2079-8999 (M. Galletti); 0000-0002-7153-6094 deployed the interface. Second, we tested whether AI
(E. Pasqua); 0000-0001-6606-200X (D. Nardi) can be a valid support tool for text comprehension from
© 2024 Copyright for this paper by its authors. Use permitted under Creative Commons License
Attribution 4.0 International (CC BY 4.0).
CEUR
ceur-ws.org
Workshop ISSN 1613-0073
Proceedings
a clinical perspective. local and global. Local processing involves the un-
derstanding of individual sentences and phrases,
while global processing focuses on integrating
2. Pedagogical and these local meanings into a coherent representa-
Psycho-linguistics Foundations tion of the entire text. Central to their model is
the concept of "coherence," where readers strive
Over the years researchers in psycho-linguistics had been to create a cohesive mental representation by link-
trying to formalise how people understand connected ing information within the text and connecting
text. Text comprehension is defined as a complex cog- it to their existing knowledge.
nitive task which involves an active process of meaning
As much as the models described may emphasize different
construction, dependent not only on the information
components as central to the comprehension process,
in the text but also on the information possessed by the
they all converge on the idea that language skills, along
reader [7]. Researchers in psycho-linguistics have strived
with semantic and inferential skills, participate in linking
to formalize the mechanisms underlying this process. As
prior knowledge to new information, creating a coherent
a result, four main theories emerged as the main ones in
and more complex representation of new meanings. In
the psycho-linguistics landscape:
designing our interface, ARTIS, we drew inspiration from
1. The Simple View of Reading by Gough and these foundational theories in psycho-linguistics.
Tunmer [8] where the decoding process and lan-
guage skills are considered central to text com-
prehension. The key idea of this model is that 3. Existing Work
comprehension, C, is defined as the product of de-
In the literature, there is one Italian systems for tele-
coding factors, D, defined as fluent word reading
rehabilitation targeting specific learning disorders and
and language comprehension, L [8].
special educational needs, i.e. RIDInet [17]. RIDInet
2. Structure Building Framework by Gerns-
focuses on text comprehension skills, particularly seman-
bacher [9, 10, 11] where text comprehension is
tic and syntactic inferential processes. Neberthless, de-
defined with the metaphor of the construction of
spite the clinical context, RIDInet lacks the integration
a building where one’s must start from the foun-
of Natural Language Processing (NLP) techniques. In
dations i.e. from the first element contained in
English, two primary systems are 3D Readers [18] and
the text, for later one integrating the new infor-
CACSR [19]. 3D Readers offer users options for verbal or
mation and enriching the departing structure in
visual strategies to enhance comprehension, with imme-
a cohesive one.
diate feedback provided. CACSR provides personalized
3. Memory-Focus Model of Sanford and Gar- instruction using various techniques like visual images
rod [12, 13] in which is central the process of and summarizing strategies, offering real-time feedback.
co-reference resolution. In fact, the key idea of Although these systems enhance reading comprehen-
this theory is that when comprehending a text, sion, they are primarily used for educational contexts
a reader must first establish a coherent interpre- and lack focus on clinical categories. Moreover, Open
tation of the text comprehended so far and must Book [20] is the only system employing NLP, mainly
therefore establish whether a given element had for children with Autism Spectrum Disorder, focusing
been discussed previously or not. on text simplification and customization functions [21].
4. Construction-Integration Model of Kintsch Its key features include text simplification through NLP
and van Dijk [14, 15, 16] focuses on how infor- techniques and rich customization functions, enabling
mation from the text, at the granularity level of users to quickly adapt the document presentation (font,
propositions, is connected to and completed by in- font size, line spacing, colors) to their preferences. It
formation stored in the long-term memory of the also provides assistive tools such as dictionaries and im-
reader. Besides literal understanding of the words ages. However, Open Books was designed for a specific
and syntax in the text, it’s essential to be able to clinical population, and it focused more on improving
gather propositions from the text, their concepts their decoding skills than reading comprehension per
and connections, and organise them into intercon- se. Finally, Systems targeting tele-rehabilitation specifi-
nected hierarchical structures. It is only with this cally for Dyslexia exist [22, 23, 24, 25], but few focus on
representation of information, integrated with reading comprehension skills.
the reader’s preexisting knowledge makes it pos-
sible to gain deep access to the content of the
text[7, 16, 15]. To reflect this view, Kintsch and
van Dijk identified two main levels of processing:
4. Target Population of ARTIS ercises that focus on co-reference structures. Finally, in
a preliminary stage, ARTIS endeavors to develop users’
ARTIS targets two distinct target populations: students ability to construct broader mental models of language
with specific reading comprehension disorders, and those by integrating textual information with their existing
identified in literature as “poor comprehenders”. knowledge [31]. Subsequent sections will delve into the
Reading Comprehension Disorder, identified in the functionalities of each level, providing detailed insights
DSM-V[26], delineates difficulty in grasping the meaning into the underlying algorithms. The initial prototype of
of text despite proficient decoding skills. This encom- the interface was extensively described in [32].
passes understanding word sequences, implicit informa-
tion (inferences), and the deeper meaning of text content.
The distinction between decoding and comprehension
5.1. Understanding Sentences
abilities is highlighted in the literature [27], which under- The interface initially presents the text at a sentence level
lines differences in cognitive processes, predictive factors, using the Spacy Sentencizer tool 1 ., allowing users to fo-
disorder characteristics, and treatment approaches. The cus on individual sentences. To aid comprehension, users
functions involved in text comprehension include lexical can listen to the text through speech synthesis powered
skills (vocabulary), inferential skills, working memory, by the Google text-to-speech API, improving accessibility
attention, and meta-cognitive control, which are distinct for those with reading difficulties. Keywords from each
from decoding skills. Even if this disorder therefore often sentence are extracted and displayed using Picture Com-
intersects with other conditions such as specific language munication Symbols (PCS) on the left side of the screen,
disorder, decoding disorder (dyslexia), intellectual dis- facilitating comprehension for individuals with literacy
ability, or memory difficulties, it can be said that specific challenges. The extraction process involves using a cus-
reading comprehension disorder is somehow indepen- tomized version of Keybert [33] and manual verification
dent of specific decoding disorder. Despite complexities by speech and language therapists. These keywords are
in classification, understanding this disorder is crucial, linked to appropriate pictograms through the Arasaac
as it often gets conflated with other clinical profiles. This API 2 , replacing real images to ensure suitability for vul-
is precisely the reason behind the beginning of this work nerable audiences.
and the reason why we involved since the beginning of
its conception speech and language therapists.
The literature defines “poor readers” or “poor com-
prehenders” students who experience specific problems
in comprehension in the face of decoding skills that are
instead within the normal range [28],[29], [30]. Distin- Figure 1: Section on understanding sentences.
guishing poor comprehension from a reading or writing
learning disability necessitates diagnostic test results. For
example, dyslexia manifests as reading effort, character- 5.2. Understanding Words
ized by decoding and reading difficulties, error preva-
In a following step, the interface offers a closer exami-
lence, and fluency deficits. In contrast, poor readers and
nation of the text at the word level. It displays the ten
writers exhibit milder characteristics, including adequate
rarest words for each selected text, identified using Word-
reading speed with occasional fluency issues. Poor com-
Freq [34], a tool providing frequency estimates across
prehenders’ difficulties often stem from factors such as
languages. This feature aims to assist individuals with
inadequate reading and writing process automation, en-
reading comprehension difficulties by addressing their
vironmental stimulation deficiencies, cognitive, memory,
hesitancy towards unfamiliar vocabulary. Upon user se-
or attention issues, rather than neurological causes.
lection, the word’s definition, along with a PCS repre-
senting its meaning and the original sentence context,
5. Architecture of ARTIS is provided. Definitions, sourced from the Oxford Dic-
tionary API 3 , are presented in both written and spoken
ARTIS is an online tool designed to enhance reading com- forms through speech synthesis, enhancing accessibility.
prehension skills. It comprises four distinct levels aimed 5.3. Understanding Paragraphs
at fostering a foundational understanding of language.
Based on insights from [6], a co-reference resolution algo-
The tool assists users in comprehending the lexical and
rithm was incorporated to aid users in grasping the con-
syntactic aspects of the text, beginning with understand-
nections between different entities and pronouns within
ing sentences and progressing to words. Furthermore, the
interface aids users in constructing coherent sequences 1 https://spacy.io/api/sentencizer
and hierarchical structures within the text through ex- 2 https://arasaac.org/
3
https://developer.oxforddictionaries.com/
Network, users explore the interconnections of words,
fostering a deeper understanding of their multifaceted
meanings.
Figure 2: Section on understanding words.
the text. Pronouns were prioritized due to the challenges
associated with their recognition. In this level of the inter- Figure 4: Sectiong on bridging existing and newer knowledge.
face, sentences are segmented, and grammatical elements
such as common nouns, proper nouns, and pronouns are
6. Empirical Evaluation
highlighted with distinct colors based on their Part of A pilot study was conducted at the CRC in Rome, a de-
Speech Tags. This color-coded approach assists in imme- velopmental rehabilitation center, from January to June
diate differentiation between elements and suggests the 2023. The study involved 24 Italian-speaking children
function of each lexical morpheme, guiding the under- from primary 2 to 5, who were assessed using the plat-
standing of pronouns by identifying their correct refer- form under the guidance of speech therapists from the
ents. Users are presented with propositions and asked CRC. Each child had a clinical diagnosis and an updated
to select the correct referent from four alternatives. One functional profile, including IQ, language, and learning
is accurately identified by the co-reference resolution profiles. The study included children with impairments
algorithm, while two are intentionally misleading, and in text comprehension and language skills, particularly
the fourth closely resembles the correct answer. Positive in receptive and expressive vocabulary not aligned with
feedback is provided upon selecting the correct answer. their age reference. The participants comprised children
Spacy part-of-speech tagger 4 was used for parsing, and diagnosed with Specific Learning Disorder (7 subjects),
a fine-tuned version of Coreferee 5 served as the corefer- Primary Language Disorder (8 subjects), Borderline In-
ence resolution algorithm. tellectual Functioning (7 subjects), and High Functioning
Autism (3 subjects).
The children who participated in the feasibility study
underwent standardised pre- and post-treatment evalu-
ations, with particular attention in the post-evaluation
to data on reading comprehension and language skills.
The speech therapists who followed the children in the
trial were also asked to fill in a questionnaire (google
Figure 3: Section on understanding paragraphs.
form) to investigate usability, functionality and the level
5.4. A first attempt for bridging Textual of perceived effectiveness of ARTIS in its different com-
Understanding with Prior Knowledge ponents, in relation to the different pathologies, with
qualitative questions structured according to a five-point
As emphasized by [31], text comprehension extends be- likert scale (Very Little, Slightly, Sufficient, Very Much).
yond its surface-level representation. To foster deeper un- The data collected focused on the platform usage, work
derstanding, Synset Networks were introduced. Each se- area preferences, exercise types, and their correlation
lected word is presented alongside its associated Synsets with different clinical profiles. The therapists’ feedback
from the Merriam Webster thesaurus, facilitating vocab- regarding exercise functionality, including the need for
ulary expansion and recall. This approach enhances com- additional support or information beyond AI-generated
prehension by elucidating the various meanings a term content, was also evaluated.
can encompass within a text and linking it with familiar Results indicated that all participants (24 children) uti-
words, thereby aiding in the integration of new infor- lized the platform to visualize sentences and aid com-
mation with existing knowledge. Through the Synset prehension of selected texts. This feature included dis-
4
https://spacy.io/usage/linguistic-features playing relevant images corresponding to individual sen-
5
https://github.com/richardpaulhudson/coreferee tences and the option for speech synthesis reading. No
supplementary material or therapist intervention was promoting inclusive hybrid speech and language therapy
required for this aspect. Children with more significant practices.
language deficits (DPL, FIL, AUTISM) tended to use the Looking ahead, future work could focus on several as-
vocabulary area (focused on words) more frequently. The pects. Firstly, conducting an extensive clinical evaluation
inclusion of definitions accompanied by pictures proved is essential to assess the effectiveness of the interface
highly beneficial. Occasionally, additional information in clinical settings. Additionally, we aim to integrate
sourced from Google searches was needed to clarify or concept-map-based document summaries, as highlighted
enhance the meaning of certain words. A small subset in [35], to enhance the bridging between knowledge
of the sample (4 subjects diagnosed with DSA) engaged present in the text read and prior knowledge.
with the semantic network. Utilizing the semantic net-
work required meta-cognitive and generalization skills,
with therapist support in tasks such as image searches, References
reflection on different word uses, and identification of
[1] J. Oakhill, Instantiation in skilled and less skilled
texts or phrases highlighting semantic distinctions.
comprehenders, The Quarterly Journal of Experi-
The results of the pre and post-treatment evaluations
mental Psychology Section A 35 (1983) 441–450.
show a significant increase in text comprehension and
[2] K. Nation, M. J. Snowling, Factors influencing
some language skills with particular reference to recep-
syntactic awareness skills in normal readers and
tive and expressive vocabulary.
poor comprehenders, Applied psycholinguistics 21
(2000) 229–241.
7. Risks and Mitigations [3] R. Padovani, La comprensione del testo scritto in
età scolare. una rassegna sullo sviluppo normale e
To ensure the development of a trustworthy AI sys- atipico, Psicologia clinica dello sviluppo 10 (2006)
tem, extensive involvement of key stakeholders, particu- 369–398.
larly speech and language pathologists, was maintained [4] K. Nation, M. J. Snowling, Beyond phonological
throughout the work. Their participation ranged from skills: Broader language skills contribute to the
the initial decision-making phases to the final evalua- development of reading, Journal of research in
tion of the interface, with continuous exchange of advice reading 27 (2004) 342–356.
and feedback. Moreover, the administration of exercises [5] F. Arosio, C. Branchini, L. Barbieri, M. T. Guasti,
always occurs under the supervision of practitioners. Failure to produce direct object clitic pronouns as a
Before conducting clinical trials involving children, pre- clinical marker of sli in school-aged italian speaking
cautionary measures were taken to mitigate risks. Texts children, Clinical linguistics & phonetics 28 (2014)
were pre-selected and categorized by school age to cater 639–663.
to users’ individual levels. Additionally, manual clini- [6] A. Zachou, E. Partesana, E. Tenca, M. T. Guasti,
cal control was implemented to identify and rectify any et al., Production and comprehension of direct
potentially unsafe pictograms or misleading keywords object clitics and definite articles by italian children
extracted by the AI models. A stop button was integrated with developmental dyslexia, Advances in language
into the interface to prevent exposure to misleading ma- acquisition (2013) 464–471.
terial. Furthermore, certain functionalities, such as the [7] W. Kintsch, On comprehending stories, Cognitive
semantic network, are selectively displayed based on in- processes in comprehension (1977) 33–62.
dividual needs and clinical judgment, ensuring tailored [8] P. B. Gough, W. E. Tunmer, Decoding, reading, and
support for each user. reading disability, Remedial and special education
7 (1986) 6–10.
[9] M. A. Gernsbacher, Language comprehension as
8. Conclusions & Future Work structure building, Psychology Press, 2013.
[10] M. A. Gernsbacher, Cognitive processes and mech-
In this paper, we described the development of ARTIS,
anisms in language comprehension: The structure
an AI-powered interface designed to enhance text com-
building framework, in: Psychology of learning and
prehension in children with language and learning disor-
motivation, volume 27, Elsevier, 1991, pp. 217–263.
ders. By leveraging neuro-psycholinguistic models and
[11] M. A. Gernsbacher, Two decades of structure build-
thanks to the integration of AI algorithms, ARTIS au-
ing, Discourse processes 23 (1997) 265–304.
tonomously extracts keywords, associates them with pic-
[12] A. J. Sanford, S. C. Garrod, S. Garrod, Understanding
tograms, identifies complex words, generates semantic
written language: Explorations of comprehension
networks, and proposes exercises on grammatical compo-
beyond the sentence, John Wiley & Sons, 1981.
nents. These features aim to improving text comprehen-
sion skills in children with special educational needs by
[13] A. J. Sanford, S. C. Garrod, Selective processing in laka, T. Edirisinghe, S. Thelijjagoda, Alexza: A
text understanding. (1994). mobile application for dyslexics utilizing artificial
[14] T. A. Van Dijk, W. Kintsch, et al., Strategies of intelligence and machine learning concepts, in:
discourse comprehension (1983). 2018 3rd International Conference on Information
[15] W. Kintsch, The role of knowledge in dis- Technology Research (ICITR), IEEE, 2018, pp. 1–6.
course comprehension: A construction-integration [26] A. P. Association, et al., Diagnostic and statistical
model, in: G. Stelmach, P. Vroon (Eds.), Text manual of mental disorders, Text revision (2000).
and Text Processing, volume 79 of Advances [27] A. P. Association, et al., Manuale diagnostico e
in Psychology, North-Holland, 1991, pp. 107– statistico dei disturbi mentali, Quinta edizione. Raf-
153. URL: https://www.sciencedirect.com/science/ faello Cortina Editore (2014).
article/pii/S0166411508615514. doi:https://doi. [28] D. V. Bishop, M. J. Snowling, Developmental
org/10.1016/S0166-4115(08)61551-4. dyslexia and specific language impairment: Same
[16] W. Kintsch, How readers construct situation mod- or different?, Psychological bulletin 130 (2004) 858.
els for stories. The role of syntactic cues and [29] J. Oakhill, K. Cain, Issues of causality in children’s
causal inferences, Typological Studies in Language reading comprehension, Reading comprehension
31, John Benjamins., Amsterdam; Philadelphia, strategies: Theories, interventions, and technolo-
1995. URL: https://www.degruyter.com/database/ gies (2007) 47–71.
COGBIB/entry/cogbib.6551/html, 2010. [30] B. Carretti, R. De Beni, C. Cornoldi, et al., Disturbi
[17] RIDInet, https://www.anastasis.it/ridinet/), ???? della comprensione del testo, in: Difficoltà e disturbi
[18] 3D Readers Software, https://www.sbir.gov/node/ dell’apprendimento, Il Mulino, 2007, pp. 143–162.
334161), ???? [31] W. Kintsch, T. A. Van Dijk, Toward a model of
[19] A.-H. Kim, S. Vaughn, J. K. Klingner, A. L. Woodruff, text comprehension and production., Psychological
C. Klein Reutebuch, K. Kouzekanani, Improving review 85 (1978) 363.
the reading comprehension of middle school stu- [32] M. Galletti, E. Pasqua, F. Bianchi, M. Calanca,
dents with disabilities through computer-assisted F. Padovani, D. Nardi, D. Tomaiuoli, A reading
collaborative strategic reading, Remedial and spe- comprehension interface for students with learning
cial education 27 (2006) 235–249. disorders, in: Companion Publication of the 25th In-
[20] E. Barbu, M. T. Martín-Valdivia, L. A. U. Lopez, ternational Conference on Multimodal Interaction,
Open book: a tool for helping asd users’ semantic 2023, pp. 282–287.
comprehension, in: Proceedings of the Workshop [33] M. Grootendorst, Keybert: Minimal keyword extrac-
on Natural Language Processing for Improving Tex- tion with bert., 2020. URL: https://doi.org/10.5281/
tual Accessibility, 2013, pp. 11–19. zenodo.4461265. doi:10.5281/zenodo.4461265.
[21] N. Pavlov, User interface for people with autism [34] R. Speer, rspeer/wordfreq: v3.0, 2022. URL: https:
spectrum disorders, Journal of Software Engineer- //doi.org/10.5281/zenodo.7199437. doi:10.5281/
ing and Applications 2014 (2014). zenodo.7199437.
[22] M. Wang, B. Muthu, C. Sivaparthipan, Smart as- [35] M. Galletti, M. Anslow, F. Bianchi, M. Calanca,
sistance to dyslexia students using artificial intelli- D. Tomaiuoli, R. Van Trijp, E. Pasqua, Interactive
gence based augmentative alternative communica- concept-map based summaries for send children,
tion, International Journal of Speech Technology 2022, pp. 5236–5243.
(2021) 1–11.
[23] A. Zingoni, J. Taborri, V. Panetti, S. Bonechi,
P. Aparicio-Martínez, S. Pinzi, G. Calabrò, Inves-
tigating issues and needs of dyslexic students at
university: Proof of concept of an artificial intelli-
gence and virtual reality-based supporting platform
and preliminary results, Applied Sciences 11 (2021)
4624.
[24] P. D. Barua, J. Vicnesh, R. Gururajan, S. L. Oh,
E. Palmer, M. M. Azizan, N. A. Kadri, U. R. Acharya,
Artificial intelligence enabled personalised assis-
tive tools to enhance education of children with
neurodevelopmental disorders—a review, Interna-
tional Journal of Environmental Research and Pub-
lic Health 19 (2022) 1192.
[25] S. Rajapakse, D. Polwattage, U. Guruge, I. Jayathi-