Application of artificial intelligence in screening of birth defects a scoping review

MINISTRY OF EDUCATION MINISTRY OF HEALTHHANOI MEDICAL UNIVERSITY ____♦♦♦____ HUY DO DUC APPLICATION OF ARTIFICIAL INTELLIGENCE IN SCREENING OF BIRTH DEFECTS: A SCOPING REVIEW GRADUATION

MINISTRY OF EDUCATION MINISTRY OF HEALTH

HANOI MEDICAL UNIVERSITY

♦♦♦

HUY DO DUC APPLICATION OF ARTIFICIAL INTELLIGENCE IN

SCREENING OF BIRTH DEFECTS: A SCOPING REVIEW

GRADUATION THESIS DOC TOR OF PREVENTIVE MEDICINE

School of Preventive Medicine and Public Health, as well as teachers from theDepartment of Epidemiology for your guidance and support.

I would like to express my great appreciation to all the individuals and teamsbelow, that without them I would not be able to accomplish my graduation thesis.Firstly I would like to express my deepest gratitude to my supervisor.Assoc.Prof Le Minh Giang for his guidance Even though he has always been busy,

he still willing to give me some of his precious time This has always been verymuch appreciated

The second and third person that I want to say thanks to is A/Prof NguyenThi Trang and Ms Le Thi Minh Phuong from Department of Biomedical Geneticsfor your help in birth defects, especially Ms Le T1Ũ Minh Phuong for the time youspent screening articles with me

In addition, because I can always ask you about scoping review, thank you

Ms Nguyen Thi Hue from Center for Research and Training on Substance HIV

Abuse-Last but not least, I would like to give thanks to my family and close friends,for their support and encouragement throughout my study

Hanoi May 2021

Do Due Huy COMMITMENT

Respectfully addressed to:

Board of Hanoi Medical University

Board of Preventive Medicine and Public Health School

Department of

Epidemiology-Board of Dissertation Assessment

My name is Do Due Huy - Student of Hanoi Medical University, course 2015

- 2021 majoring in Preventive Medicine Doctor, hereby declare that:

This is a research that I conducted under the scientific guidance of Assoc.Prof

Le Minh Giang The data and results presented in the research are completelytruthful In addition, die thesis also uses a number of comments, assessments aswell as results from other authors, agencies and organizations, all with sourceannotations clearly stated in the references

I will take full responsibility if tliere was any' fraud in the contents of myresearch

Hanoi May 2021

Do Due Huy

ABBREVIATIONS

TABLE OF CONTENT

ABSTRACT

INTRODUCTION 1

LITERATURE REVIEW 3

2.1 Scoping review • ••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••é 3 2.2 Birth defects - - - - -4

2.2.1 Definition 4

2.2.2 Classification of birth defects 4

2.23 Cause of birth defects in human 5

2.2.5 Screening methods for some common birth defects 9

2.3 Artificial Intelligence (Al) 15

2.3.1 Al application in health care 15

2.3.2 Definition - - 16

2.3.3 Machine learning 2.33 Evaluate the effectiveness of artificial intelligence software 19

17

••••••••• 21

3.8 Synthesis of results 24

RE SITT 25

4.1 Results of articles search 4.2 Characteristics of studies - 26

4.3 Al development and effectiveness — 26

DISCISSION 31

CONCLUSION 42

RECOMMENDATIONS 43

REFERENCES _44 APPENDLX 49 3.1 Protocol and registration 21

3.2 Study subjects 21 3.2.1 Inclusion criteria 21 3.2.2 Exclusion criteria 21 3.3 Information sources 22 3.4 Search 22 3.5 Selection of sources of evidence - 23

3.6 Data charting 23 3.7 Data Items •• •• • ••••••••••• •• •••••••••••• •• • •• • • • ••••••••• •• • •••••••••• • ••••••• • • • • • • • •• •• • •••••••• •

25

LIST OF TABLES

Table 2.1 The risk of Down Edward and Palau Syndrome according to maternal

age 6

Table 2.2 Diagnosis of abnormalities using AFP uE3 and HCG 13

Table 2.3 Confusion matrix 20

Table 4.1: General characteristics of eligible studies 26

Table 4.2 Methodological characteristics of eligible studies 27

Table 4.3: Methodological characteristics of eligible studies 30

Table Ô.4 Limitation in eligible studies 39

LIST OF FIGURES

Figure 2.1 Maternal age-related risk for trisomy 21 at 12 weeks gestation and

maternal serum b-hCG levels (left) and PAPP-A (right/ 12

Figure 2.2 Example of an supervised learning system 19

Figure 4.1: Literature search and study identification strategy 25

Figure 5 1 An example of support vector machine 34

Figure 5.2 An illustration of neural network 35

ABSTRACT

Introduction: There are many advantages of artificial intelligence in healthcaresetting compared to human clinicians However, the readiness of Al to replacehuman as a stand alone screening program is remain unknown

Method: We conducted a scoping review, a structured evidence synthesisdescribing a broad research field, to summarise knowledge on Al evaluated for birthdefects screening and to assess Al’s ability for adoption in birth defects screening in2010-2020 period Data were collected through PubNied database using acombination of keywords which was agreed by three different reseachers

Result: There were 11 eligible studies All Al models built in these studiesarchieved encouraging results However, there are still key evidence gaps that need

to be addressed before Al can be rendered more transferable to large- scalescreening evaluations

Conclusion: We found that die published evidence on Al application for birthdefects detection was concentrated around model (algorithmic) development,generally independently of real-world clinical or screening evaluation, and overallthe evidence does not indicate readiness of Al systems for real-world birth defectsscreening trials

Keywords: Artificial Intelligence; Birth Defects; Screening

There are about 303.000 newborns who died within 4 weeks of birth everyyear, worldwide, due to congenital abnomalies* In the vs for every 33 babies bom.there is 1 with a birth defect*' Birth defects can have lasting effects, withdevastating consequences not only for the child, but also for the family, the healthsystem, and society as a whole* Early screening methods for intervention andtreatment can be implemented to limit the complications of birth defects in order toimprove quality of life for children and families as well as reducing infant mortalitydue to birth defects3 Therefore, we need to have effective methods to screen,diagnose and predict birth defects so that we can ha\e early interventions

In recent years, artificial intelligence (Al) has become an important part of ourdaily life Some applications of artificial intelligence in everyday life includeGoogle Translate" Google Map5 Youtube video proposal software6 Inhealthcare, artificial intelligence lias many important applications such as screeningfor cancer or building predictive models in disease prevention Application ofartificial intelligence to build clinical decision support system is of interested tomany scientists around the globe3 The advantages of Al includes promotingevidence-based diagnostics, improving diagnostic efficiency, individualizingtreatment as well as bringing high economic efficiency9,10

With these advantages, the application of artificial intelligence in prenatalscreening can be an effective tool to help central hospitals to make more accuratediagnosis of prenatal birth defects as well as replacing doctors in lower levelhospitals where there are no trained specialists Therefore, it is necessary' to assessthe ability' of artificial intelligence systems to screen for birth defects based onexisting literatures and identify' gaps in these studies However, there are not manystudies mentioned this issue Thus, we conducted tills study with 2 objectives:

Objective 1: To describe characteristics of current studies in the 2010 - 2020period on the application of artificial intelligence in screening of birth defects inpregnant women.

Objective 2: To assess Al's readiness in large scale screening of birth defects

The term "scoping review" is an ideal means of determining the coverage of adocument on a particular issue and giving clear indication of the number of studiesavailable as well as generalizing the concentration of studies to that problem15'*4 Althoughsimilar to systematic review that scoping review follow a pre-built process, they are donefor different purposes1’ It can be used to synthesize concepts about a particular area ofstudy, along with clarifying the definitions or conceptual boundaries of a topic10 Alongwith that, the scoping overview can be the foundation for a systematic review study,especially in the case when we need to learn about a new topic but there are still a lot ofunanswered questions1 We can distinguish between scoping reviews and systematicreviews based on research questions If the authors have a question regarding thefeasibility, appropriateness, meaningfulness or effectiveness of a certain problem, thensystematic renew is likely the most valid approach15 However, if they don't ask suchprecise question, and want to explore the certain characteristics/ concepts in previousstudies, or simply reporting and discussing about these problems, then scoping review is abetter choice.

2.2 Birth defects

2.2.1 Definition

Congenital anomaly (Birth defect or Congenital Disorder or CongenitalMalformation), is defined by many authors using different factors depending on theirpurposes, but they all agree on the following points:

- All are abnormalities that have a prenatal cause

- These abnormalities may manifest at the bodily, cellular or molecular level

- These abnormalities manifest at birth or in later stages

Thus, birth defects are all abnormalities at the bodily, cellular or molecular level,which can manifest at birth or at a later stages but have a prenatal cause.19

2.22 Classification of birth defects

There are many ways to classify birth defects, but we introduce a classificationsystem based on ICD- 10:

- Q0O-QO7 Congenital malformations of the nervous system

- Q10-Q18 Congenital malformations of eye ear face, and neck

- Ọ20-Q28 Congenital malformations of the circulatory system

- Q30Q34 Congenital malformations of the respiratory system

- Q35-Q37 Cleft lip and cleft palate

- Q38-Q45 Other congenital malformations of the digestive system

- QS0-Q56 Congenital malformations of genital organs

- Q60-Q64 Congenital malformations of the urinary system

- Q65-Q79 Congenital malformations and deformations of the musculoskeletal system

- Q80 Ọ89 Other congenital malformations

- Q90-Ọ99 Chromosomal abnormalities, not elsewhere classified

2.23 Cause of birth defects in human

Causes of birth defects are divided into 3 groups by many authors, including: (1)caused by genetic factors (2) caused by environmental factors, (3) caused by multiple

genetic factors In addition, there are many birth defects with unknown causes.

a Birth defects caused by genetic factors

Birth defects caused by genetic factors are divided into two groups which are birthdefects caused by chromosome disorders and birth defects caused by single gene mutations

- Birth defects caused by chromosome disorders:

Chromosomal disorders are changes in chromosome number, structure, or mosaic,resulting in the addition or loss of genetic material In this group, aneuploidy ofchromosome 21 (trisomy 21), also known as Down’s syndrome, was the most common Inaddition, there are some other common birth defects such as Patau syndrome Edwards.Turner Klinefelter

- Birth defects caused by single gen mutations:

There are many single gene mutations that cause the genetic birth defects and itsexpression obeys the genetic law's of Menden Single-gene diseases have 3 types of geneticmechanisms which are autosomal dominant, recessive and sex-linked There are more than6.000 types of single-gene mutations that have been described20, of which brittle Xchromosome syndrome is the most common genetic cause of mental retretness In addition,there are also defects such as skeletal dysplasia, cartilage dysplasia, cartilage hypoplasia,microcephaly, colorblindness, hemophilia A

b Birth defects caused by environmental factors

at liigh risk of environmental exposures which are harmful to the fetus such as tobacco,alcohol, etc In addition, there are still may limitations in reproductive health care services,family planning or prevention of congenital syphilis syndrome, congenital rubella in thehealth care system which leads to high proportion of birth defects22.

- Nutrition deficiencies:

A nutrition deficiency diet during pregnancy can affect the risk of having a child withbirth defects Lack of Phosphorus, Magnesium and other trace elements can lead to adeformation of the skeleton, causing congenital rickets22

An iodine deficient diet during pregnancy can also give birth to a baby uiwth birthdefects UNICEF believes that iodine deficiency causes brain damage and intellectualdisability20 However, this is also the leading cause of mental retardation which can easily

be prevented20,25 Iodine Deficiency Disorder causes spontaneous abortion, perinatal death,mental retardation, hearing impairment, etc Severe iodine deficiency disorder can lead to

Folic acid is a vitamine (vitamin B9) required for biosynthesis and methylation ofDNA and RNA It is very' important for cell division, especially at the time of rapid celldivision such as in an embryo Folic acid is essential for the development of tile brain andspinal cord during the first 4

weeks of pregnancy24 Reports indicate that 95% of babies with neural tube defects occur inmothers with no family history of the disease There are extensive scientific evidenceslinking birth defects with folate deficiency’ and the use of folic acid before conception withprevention of neural tube defects24

- Physical and chemical agents:

Chemical agents such as pesticides, plant protectors, heavy metals such as mercury,lead can cause birth defects such as eye defects, limb deformities, facial and mouthdefects Dioxin can cause birth defects at 2 times higher rate than other agents25

Physical agents such as radioactive substances can influence embryogenesis whileradiation, gamma rays, and ultraviolet rays can also disrupt polymorphic process of muscleand other organs of the embryo

- Microbiological agents:

Mother infected with viruses during pregnancy such as cytomegalovirus, zoster virus, chickenpox, flu especially rubella can give birth with birth defects such asdefects of the nervous system, cardiovascular system, calcification in the brain,microcephaly, mental retardation

herpes-Some bacteria such as toxoplasma, chlammydia trachomatis, syphilis can alsocause birth defects such as fetal death, cleft lip calcification in the brain, hydrocephalus,encephalitis - meningitis, retinitis if the mother has been previously infected withouttreatment or infected during pregnancy20

- Drinking alcohol during pregnancy:

Fetal Alcohol Syndrome (FAS) is a syndrome of birth defects in children whosemothers drink alcohol during pregnancy This syndrome includes growth retardation, heartdefects, physical, mental and behavioral disorders that can include low IQ or mentalretardation Fetal alcohol syndrome is not a simple birth defect It is a group or pattern of

related disorders The severity of symptoms varies, with some children having worsesymptoms than others Alcohol can affect the fetal brain at any time during pregnancy.Hence there is no safe alcohol dosage, no safe drinking time or no right kind of wine todrink during pregnancy2'

- Some other factors:

Some other factors such as obesity mother, insulin-dependent diabetes, usage ofstimulants such as cocaine, smoking, sedatives (Thalidomide ), antiepileptic drugs arealso risk factors that increases the rate of childbirth with birth defects

2.25 Screening methods for some common birth defects

a Ultrasound

Ultrasound is a non-invasive procedure that does not harm both the mother and thefems, which allows clinicians gather some information about the pregnancy that cannot beprovided by any examination such as: gestational age number of fetuses, fetaldevelopment, mother-to-child metabolism quality(based on Doppler) and fetal morphology.Although there have been many technical improvements, but ulttasound is still not theperfect method, it can only detect some fetal malformations when the fetus is in a favorableposition with the right amount of amniotic fluid Unclear morphological abnormalities arealso difficult to detect on ultrasound For example, in Down Syndrome, ultrasound can onlydetect indirect images such as nuchal translucency

Ultrasound for tile measurement of nuchal translucency is usually done at 11-13weeks of pregnancy which will give the most accurate results The majority of cases withnuclial translucency < 3 mm were classified as low-risk (less likely to developchromosomal abnormalities) In the case when nuchal translucency is ranged from 3.5 to4.4 mm there is a chromosomal abnormality rate of 21.1% and in the case which it is > 6.5

mm the risk of chromosomal abnormality can be increased up to 64.5% In cases wherenuchal translucency is > 3 mm the pregnant woman will be ordered to perform an

additional triple test at 16-18 weeks

b Double test Triple test

The first Down syndrome screening method was imroduced in the 1970s based onmaternal age women over 40 years old will be given an amniocentesis test to determine therisk of fetus with chromosomal abnormalities Later, when amniocentesis becomes saferthan before with the guidance of ultrasound the cost is also reduced, amniocentesis iswidely indicated in high-risk pregnant women, ie older than or equal to 35 years old

Test of biochemical indices in maternal blood

- Maternal serum alplia-fetoprotein (AFP)

A developing fetus lias 2 main types of blood protein Albumin and alpha fetoprotein(AFP) while an adult lias only albumin, so an AFP test in the maternal serum is used toindirectly determine the amount of AFP in the fetal blood

Normally only a small amount of AFP in the amniotic fluid can cross the placenta toenter the mother's bloodstream However, when there is a neural tube abnormality, becausepart of The embryonic neural tube is not closed, AFP will escape into the amniotic fluid.Neural tube abnormalities include anencephaly (due to the neural tube that does not closethe head) and spina bifida (due to the inability of the tail of the neural tube) In the US therate of these diseases is 1-2/1000 births Likewise in gastroschisis or omphalocele, AFPfrom the fetus enters mother's bloodstream in a larger amount than usual

AFP tends to be lower than normal in fetuses with Down syndrome or somechromosomal abnormalities, so AFP is useful in screening for Down syndrome and anumber of other infections A combination of AFP screening and ultrasound can detectalmost all anencephaly and most cases of spina bifida

- Maternal serum free Beta-HCG

This is the most commonly used test during pregnancy About 1 week after theembryo implants in the uterus, the amount of beta HCG secreted by the culturing cells is

sufficient to diagnose pregnancy In the earl}’ stages of pregnancy, beta HCG helps in earlydiagnosis and prognosis of miscarriage, ectopic pregnancy because in these cases, betaHCG is lower than normal

Later in pregnancy, at the end of the second trimester HCG may be used incombination with AFP to screen for specific chromosomal abnormalities in Downsyndrome Increased HCG in association with decreased AFP is an implication of Downsyndrome Meanwhile, abnormally high hCG suggests pseudocyesis

Figure 2.1 Maternal age-related risk for trisomy 21 at 12 weeks gestation and

maternal serum b-hCG levels deft) and PAPP-A (right)

- Maternal serum Estriol

Estriol is derived from dehydroepiandrosterone (DHEA) which is produced from theadrenal glands and then converted to estriol by the placenta Estriol enters the mother'sbloodstream and is excreted in the urinary tract or excreted by the liver into the bile.Continuous testing of estriol in the third trimester is performed to monitor fetal healthstatus If the concentration of estriol is reduced, tile fetus is at risk and may indicate an end

to pregnancy Estriol is also reduced in fetuses with Down syndrome or adrenalinsufficiency or anencephaly

- Pregnancy-associated plasma protein A (PAPP-A)

In the first trimester, low serum PAPP-A is an indication of trisomies 13,

1 s and 21 Furthermore, low PAPP-A levels in the fust trimester predict a low

4.0 MoM 3.0 MoM

10 MoM link ground

0.25 MoiM

<>35 MoM 0.50 MoM

B M kgnxmd

is inhibin-A.

Table 2.2 Diagnosis of abnormalities using AFP, uE3 and HCG

(The values of these indicators depend on gestational age)

c Amniocentesis

Amniocentesis is the most widely used method today because of its technicalsimplicity as well as low rate of complications It is considered the main method ofobtaining fetal specimens

Amniocentesis is done at 3 periods: Early amniocentesis (13 to 16 weeks gestation),classic amniocentesis (from 17 to 20 weeks gestation), late amniocentesis (after 20 weeks ).The best gestational age for this procedure is 17 to 18 weeks because at this time thechance to successfully draw out amniotic fluid is highest while the rate of complications forboth mother and fetus is lowest The procedure is performed under ultrasound guidance.Amniocentesis takes about 20 ml of amniotic fluid for testing The whole proceduretakes 5 to 10 minutes Then maternal need to stay in place for 3 hours, she doesn't have touse antibiotics Risk of amniotic fluid leakage, miscarriage is < 0.5% Test results will be

available in 2 to 3 weeks

- Chorionic Villus Sampling

Chorionic villus sampling (CVS), or chorionic villus biopsy, is a prenatal test thatinvolves taking a sample of tissue from the placenta to test for chromosomal abnormalitiesand certain other genetic problems T1ŨS method causes high rate of miscarriage (about9%), so it is only used mainly in cases of fetus with severe abnormalities detected in thefirst trimester This method is performed under ultrasound guidance Results will beavailable after 5 to 7 days

- Non-invasive prenatal screening (NIPT)

This is considered to be the most effective and safest testing method available today.The method is performed early from the 10th week of pregnancy through the mother'sblood sample (only 7-10 ml) Chromosome abnormalities can be screened includechromosome 6, 9 13 (Palau's syndrome), chromosome 18 (Edwards), chromosome 21(Down), chromosome X Y and segmental mutations, etc In addition Illis method is alsoapplicable for single pregnancy, twins surrogacy with high aocuracv, up to 99.98%

Prenatal screening not only helps detect birth defects for the fetus, but it also enhancesand improves the quality of fiiture generations Therefore, prenatal screening and diagnosis

is an essential job to help mothers detect diseases, have healthy babies which developnormally

At the present, modern molecular biology- techniques such as NGS, microarray.BoBs can screen and identify- many genetic mutations Especially the NGS technique candecodes the entire human genome to identify' genetic abnormalities related to birth defects.This technology help US detect 4500 different genetic diseases Therefore, it is necessaty tohave machine learning and deep learning technology in order to promote the effectiveness

of these systems

2.3 Artificial Intelligence (.Al)

2.3.1 Al application in health care

In clinical setting Al is often used to build clinical decision support system Clinicaldecision support systems haw the responsibility of assisting physicians or healthcareprofessionals in making clinical decisions These systems can improve the quality ofhealthcare by providing references to doctors based on data from the past

Clinical decision support systems have been developed since the 1970s, such as theMYCIN INTERNIST-1 and CASNET Decision support systems have been developed andapplied in many fields of healthcare such as diagnosing diseases, developing treatmentregimens, making drugs, monitoring and taking care of patients These systems have beenapplied in many facilities around the world Al algorithms in healthcare have beendeveloped by many companies, including large companies such as IBM Microsoft.Google Intel Facebook and even startups In recent years, the development and application

of clinical decision support systems have been increasingly strengthened There are manyreasons for this development, that is the development of hardware leading to increasedcomputing power shortening The time to collect and process data; the volume of medical-related data collected from medical and personal devices is increasing; development ofgenetic databases: electronic medical record management systems are becoming more andmore popular: development of highly accurate advanced Al techniques (such as deeplearning methods) in the fields of computer vision and natural language processingenhances the accuracy of Al system 2.32 Definition

Artificial Intelligence (Al), also known as Al is the intelligence expressed bymachines, different from the natural intelligence expressed by humans or animals, whichare related to consciousness and emotions

In other words, artificial intelligence is a branch of computer science whose purpose

is to give software the ability.- to analyze information, then make decisions based on

results

The artificial intelligence concept is built based on 2 different ideas The first idea isaritificial narrow intelligence (also known as ANI) These are Al softwares that can do aspecific job like voice recognition or selfdriving cars or sorting spam messages Tiresecond idea is about artificial general intelligence (artificial general intelligence), alsoknown as AGI This is the purpose of building artificial intelligence, which is to createsoftware that can do everything humans can or even more Today, almost all new Alinventions belong to ANI

Any machine learning method can be categorized into one of two types: supervisedlearning or unsupervised learning

a Supervỉsỉed learning

As children, we learned to classify new things under guidance of adults They point at

a furry four leg creature that bark and tell US that it is a dog Through many suchinstructions, we get to know how to identify a dog among other animals This is the coreconcept of supervised machine leaning In this method, the software is given a data set andalready know how correct output should look like, having the idea tliat there is a

relationship between the input and the output The task of a supervised machine learningmechanism is to try to find the relationship between the input data and the desired output,and then use that relationship to predict the output for the new input

The supervised machine learning method is divided into two types of problems,regression and classification In a regression problem we are trying to predict results within

a continuous output, meaning that we are trying to map input variables to some continuousfunction, for example, forecasting prices for a specific house from input data such as area,number of floors, etc In a classification problem, we are instead trying to predict results in

a discrete output In other words, we are trying to map input variables into discretecategories For example, pr edicting if this person lias cancer based on a series of pictures

Tire advantage of the supervised machine learning method is that it can find out thecorrelation between input and output data that is close to reality and has good coverage ofdifferent cases However, the disadvantage of machine learning methods is that large inputdata is required, and the data must be pre-labeled, which can be expensive in terms of timeand money In addition, in order to have a good coverage of the different cases, the inputdata must be diverse and need to be updated continuously, because although it is called acat the cats in different places are different in shape, size, color, and sometimes we have toask ourselves if it’s a cat, which is the same in this machine learning approach

up with rules of a game based on observations.

The advantage of the unsupervised machine learning approach is that it does not needlabeled data, and it can provide unknown information from the input data, as well asautomatically classify' the data by finding different characteristics from the data itself.However, this method lias disadvantages such as it takes many steps to build, and it isdifficult to understand what is going on inside the software or what method it is using tolearn

2.33 Evaluate the effectiveness of artificial intelligence software

In order to classify' whether a pregnant woman has fetus with birth defects, we have

to answer a yes or no question, or in other words, positive or negative There are 4possibilities when comparing the software's prediction with fetus’s true condition If theprediction says that this case is positive and in fact this person is positive, this is called atrue positive, but if in fact this person is negative, it is called a false positive Conversely,true negative occurs when both the prediction and fetus's true condition are negative, andfalse negative occurs when the prediction is negative when in fact it is positive We candraw this table from the explanations above:

Fetus's true condition

Table 2.3 Confusion matrix

The sensitivity (sometimes also named the detection rate in a clinical setting) of thesoftware is the proponion of fetuses which test positive for birth defects among those whichtruly have the condition Mathematically, this can be expressed as:

Sensitivity’ = I True positive/ I Condition positive

Specificity of the software is the proportion of fetuses which test negative for birthdefects among those which truly do not have the condition Mathematically, this can also

be written as:

Specificity = E True negative/ E Condition negative

Finally, accuracy’ is the combination of true positive and true negative casesamong the total population Mathematically, tills can also be written as:

Accuracy’ = (E True positive + I Tme negative)/1 Total population28

MATERIALS AND METHODS

3.1 Protocol and registration

This study follows the PRISMA extension for scoping reviews was published in 2018

of Tricco The checklist contains 20 essential reporting items and 2 optional items toinclude when completing a scoping review'9 The detail checklist is showed in Appendix

3.2 Study subjects

3.2.1 Inclusion criteria

- Articles published in English, in 2010-2020 period

- Studies must show evidence of using machine learning methods applied toscreening of birth defects

- The purpose of those studies is to evaluate approaches towaid application ofartificial intelligence in the screening of birth defects

- Studies must report clinical results of Al software in screening of birth defectsthrough AUC or accuracy or sensitivity or specificity

- Studies conducted using any types of design on any group of pregnant women.3.22 Exclusion criteria

- Studies aimed to identify- markers but not the defects and their golden standards

- Studies that have animal subject

- Commentary articles, editorial articles, review articles and congress abstracts.3.3 Information sources

We searched for articles on Pubmed because it is a free resource database whichcontains more than 32 million citations and abstracts of biomedical literature There areothers popular database for literature review studies such as Scopus Web of Science, andGoogle Scholar Despite that they provide additional literatures which can cover moreaspects for our problem Scopus and Web of Science are not free databases while GoogleScholar doesn’t have an appropriate search engine for large scale articles retrieval andanalysis as well as different options for various research purposes, hence offers results ofinconsistent accuracy

3.4 Search

The search strategy was discussed by all members in our team Two reseachersindependently searched for articles using different combinations of keywords, then we