Detection Of Sickle Cell Disease Research Articles

Detection of sickle cell disease using deep neural networks and explainable artificial intelligence

Abstract Sickle cell disease (SCD), a blood disorder that transforms the shape of red blood cells into a distinctive sickle form, is a major concern as it not only compromises the blood’s oxygen-carrying capacity but also poses significant health risks, ranging from weakness to paralysis and, in severe cases, even fatality. This condition not only underscores the pressing need for innovative solutions but also encapsulates the broader challenges faced by medical professionals, including delayed treatment, protracted processes, and the potential for subjective errors in diagnosis and classification. Consequently, the application of artificial intelligence (AI) in healthcare has emerged as a transformative force, inspiring multidisciplinary efforts to overcome the complexities associated with SCD and enhance diagnostic accuracy and treatment outcomes. The use of transfer learning helps to extract features from the input dataset and give an accurate prediction. We analyse and compare the performance parameters of three distinct models for this purpose: GoogLeNet, ResNet18, and ResNet50. The best results were shown by the ResNet50 model, with an accuracy of 94.90%. Explainable AI is the best approach for transparency and confirmation of the predictions made by the classifiers. This research utilizes Grad-CAM to interpret and make the models more reliable. Therefore, this specific approach benefits pathologists through its speed, precision, and accuracy of classification of sickle cells.

Comparison of Deep Learning Techniques in Detection of Sickle Cell Disease

Recently, transfer learning technique has proved to be powerful in enhancing development of deep learning methods for sickle cell disease (SCD) detection as a complement to the clinical method where a hemoglobin electrophoresis machine is used. This is evidenced by a number of models and algorithms with ≥90% prediction accuracy. From literature, most of the proposed methods are trained and tested on pre-trained deep learning models like VGG16, VGG19, ResNet, Inception_V3 and ReNet. However, training and testing of these methods are limited on one model and separate dataset which may lead to biased results due to implementation in variation of these models which affects results produced. To this end, there exists a need to evaluate the SCD models using the same dataset. Thus, in this research study, we carried out a comparative investigation and evaluated predominate pre-trained models used to detect SCD using the same dataset to ascertain which one has the best accuracy. We used secondary dataset obtained from an online dataset. In our study, we have discovered that Inception V3 yielded the highest accuracy of 97.3% followed by VGG19 at 97.0%, VGG16 at 91%, ResNet50 at 82% and ReNet at 67%, and the CNN-scratch model achieved 81% accuracy. Results from our study will aid researchers and industry practitioners to make decision on the best deep learning model to use while detecting SCD.

Pattern of Haemoglobin Phenotypes and the Clinical Identity of Clients Tested at a Private Diagnostic Center in South-West Nigeria

Background: Haemoglobinopathies are diseases of global importance and the countries with high disease burdens have inaccurate national data on the incidence of the various haemoglobin (Hb) phenotypes.
 Objective: To review the Hb phenotype distribution, the referral identity of patients, and the clinical reasons for phenotyping.
 Methods: A retrospective descriptive study conducted at a private diagnostic laboratory in Lagos, South-West, Nigeria. The anonymous results of Hb phenotypes of the patients determined using gel electrophoresis were retrieved from the laboratory information system (LIS).
 Results: There were 942 patients: 519(55.1%) males and 423(44.9%) females with median ages 30yrs (11months-89yrs) vs 27yrs (9months-89yrs), p=0.0018. The phenotypes were HbAA, 592(63.2%); HbAS(26.4%); HbSS(7.1%); HbAC(2.1%), HbSC(0.7%) and HbCC(0.4%). Of the total, 757(80.4%) were referred by health care providers and 185(19.6%) were selfreferred, and there was no association between gender and referral identity of patients. However, there were more males [353(46.6%)] with clinical reasons for referrals than females [94(12.4%)], p<0.0001, and annual check was the commonest reason for Hb phenotyping.
 Conclusion: Most patients were referred for Hb phenotyping and an annual health check is the commonest reason for phenotype requests. This emphasizes the need for policies to support national Hb phenotype/genotype screening programmes to aid the early detection of sickle cell disease.
 Keywords: Haemoglobin phenotype; clinical identity; private laboratory.

Techniques for the Detection of Sickle Cell Disease: A Review.

Sickle cell disease (SCD) is a widespread disease caused by a mutation in the beta-globin gene that leads to the production of abnormal hemoglobin called hemoglobin S. The inheritance of the mutation could be homozygous or heterozygous combined with another hemoglobin mutation. SCD can be characterized by the presence of dense, sickled cells that causes hemolysis of blood cells, anemia, painful episodes, organ damage, and in some cases death. Early detection of SCD can help to reduce the mortality and manage the disease effectively. Therefore, different techniques have been developed to detect the sickle cell disease and the carrier states with high sensitivity and specificity. These techniques can be screening tests such as complete blood count, peripheral blood smears, and sickling test; confirmatory tests such as hemoglobin separation techniques; and genetic tests, which are more expensive and need to be done in centralized labs by highly skilled personnel. However, advanced portable point of care techniques have been developed to provide a low-cost, simple, and user-friendly device for detecting SCD, for instance coupling solubility tests with portable devices, using smartphone microscopic classifications, image processing techniques, rapid immunoassays, and sensor-based platforms. This review provides an overview of the current and emerging techniques for sickle cell disease detection and highlights the different potential methods that could be applied to help the early diagnosis of SCD.

Feasibility Study of the "HemoTypeSC" Test for the Rapid Screening of Sickle Cell Disease in Côte D'Ivoire.

Sickle cell disease is a hereditary disease that predominantly affects black people. It is very widespread in sub-Saharan Africa, particularly at the Lehmann “sickle belt” level, where the prevalence of the hemoglobin S involves at least 10% of the population in West Africa and can reach 40% in Central Africa. In Côte d'Ivoire, the prevalence of the hemoglobin S is about 12–14% in the general population and about 11.71% in the child population in Abidjan. On the other hand, its coexistence with other hemoglobin phenotypes such as AC (6.2%) and β-thalassemia (2.7%) traits may also cause composite heterogeneous sickle cell disease, e.g., SC or S/β-thalassemia in this study. Since 2009, sickle cell disease has been recognized as a public health problem; however, much still remains to be performed despite the progress achieved. The objective of this study is thus to promote a rapid screening for the struggling against sickle cell disease in Côte d'Ivoire. This study was carried out over 6 months (April–September 2019) and has included 336 children, of which 236 all-comers, recruited in the municipality of Treichville in Abidjan and 100 other children with already known hemoglobin phenotype followed up in the Hematology Department of the University Hospital of Treichville. Two tests were used: the HemoTypeSC™ for rapid screening and the hemoglobin electrophoresis which is the reference method used for confirming the diagnosis in the laboratory. The findings confirmed the reliability of the HemoTypeSC™ with a sensitivity and specificity at 100% for the detection of hemoglobin A, S, and C. On the other hand, this sensitivity and specificity drop to 98.2% and 99.7%, respectively, when we analyze all the 336 children together, including the cases with HbF detected by hemoglobin electrophoresis. Hence, the importance of performing certainty tests following the HemoTypeSC™ screening test in order to determine the accurate phenotypes and proportions of the types of hemoglobin. The prevalence of hemoglobin S in subgroup 1 of 236 children of all-comers was 15%. The HemoTypeSC™ is therefore reliable, inexpensive, and disposable for rapid screening and early detection of sickle cell disease in Côte d'Ivoire. The HemoTypeSC™ provides rapid detection of hemoglobin phenotypes HbAA, HbSS, HbSC, HbCC, HbAS, and HbAC.

A Review of Automated Methods for the Detection of Sickle Cell Disease.

Detection of sickle cell disease is a crucial job in medical image analysis. It emphasizes elaborate analysis of proper disease diagnosis after accurate detection followed by a classification of irregularities, which plays a vital role in the sickle cell disease diagnosis, treatment planning, and treatment outcome evaluation. Proper segmentation of complex cell clusters makes sickle cell detection more accurate and robust. Cell morphology has a key role in the detection of the sickle cell because the shapes of the normal blood cell and sickle cell differ significantly. This review emphasizes state-of-the-art methods and recent advances in detection, segmentation, and classification of sickle cell disease. We discuss key challenges encountered during the segmentation of overlapping blood cells. Moreover, standard validation measures that have been employed to yield performance analysis of various methods are also discussed. The methodologies and experiments in this review will be useful to further research and work in this area.

Newborn Screening for Sickle Cell Disease: Jamaican Experience.

To review the history of newborn screening for sickle cell disease with especial reference to Jamaica. A summary was done of the history, the development of associated laboratory technology and the implementation of newborn screening for sickle cell disease in Jamaica. Screening was initiated at Victoria Jubilee Hospital, Kingston from 1973-1981, reactivated in 1995 and extended to the University Hospital of the West Indies in 1997 and to Spanish Town Hospital in 1998. From August 2008, there was a progressive recruitment of 12 hospitals in the south and west of Jamaica which has raised the frequency of islandwide newborn coverage from 25% in 1973 to 81%. The results of this extended programme in southwest Jamaica are presented. Dried blood spots collected from the umbilical cord proved stable, cheap and efficient; mean sample collection rates were 98%, maternal contamination occurred in < 1% and caused diagnostic confusion in < 0.1%. By March 31, 2015, a total of 54 566 births have been screened, detecting 161 with homozygous sickle cell (SS) disease, 125 with sickle cell-haemoglobin C (SC) disease and 36 with sickle cell-beta thalassaemia. Of the 327 babies with clinically significant sickle cell syndromes, all except five who died within seven days of birth were confirmed by four to six weeks and recruited to local sickle cell clinics. Early detection of sickle cell disease and recruitment to clinics is known to reduce its morbidity and mortality. The methods currently detailed provide an effective and economic model of newborn screening which may be of value elsewhere.

The Chhattisgarh state screening programme for the sickle cell gene: a cost-effective approach to a public health problem.

In India, the Chhattisgarh State screening programme for sickle haemoglobin focuses on children aged 3-15years and has screened over 1,050,440 subjects over the last 6years. Commencing in the District around the capital Raipur, this programme has now completed screening in 7 of the 27 Districts of Chhattisgarh State. Screening is initially performed by solubility tests on fingerprick samples in the field and those with positive tests have venipunctures for haemoglobin electrophoresis. The frequency of the sickle cell trait was 9.64% and of the SS phenotype 0.29% with only two Districts in Hardy-Weinberg equilibrium, most Districts showing an excess of the SS 'phenotype' most readily explained by symptomatic selection. The estimated costs were US$0.28 (solubility tests alone) and US$0.60 (haemoglobin electrophoresis). Of the social groupings commonly used in India, the OBC's (other backward classes) had the highest frequencies of the sickle cell gene mutations, followed by the Scheduled Tribes and the Scheduled Castes. The objectives of the programme were the detection of sickle cell disease for prospective clinical management and of the sickle cell trait for purposes of genetic counselling. The former objective is being met for diagnosis although the success of referral to clinic services requires audit. The objective of genetic counselling is compromised by the failure of the screening test to detect other genes of potential clinical significance such as HbD Punjab and the beta thalassaemia trait. Despite these exceptions, the detection of HbS appears relatively robust and could be another condition factored into the traditions of partner selection amongst the underprivileged communities of this state. Overall, the Chhattisgarh programme seeks to address the daunting challenges of large populations carrying the sickle cell gene and maybe a useful model for elsewhere.

Neonatal screening for sickle cell disease and other hemoglobinopathies in "the changing Europe".

Inherited hemoglobin disorders (sickle-cell disorders and thalassaemias) represent an increasing global health problem. The early detection of sickle cell disease allows counselling for family members about disease management and future reproductive decisions. The aim of the present study was to estimate the birth prevalence of hemoglobinopathies in newborns of Italian couples and couples of immigrants from endemic areas living in an urban area of northern Italy in order to assess the opportunity of implementing a neonatal screening programme for hemoglobin disorders. Inclusion criteria were infants with at least one of the parents from high risk areas of hemoglobinopathies (Po delta and Sardinia, Italy; Mediterranean area; sub-Saharan Africa; Brazil; Asia) or a positive family history for hemoglobinopathies. The number of infants included in the present study was 337: 13.8% out of 2447 children born at Azienda Ospedaliera Universitaria (AOU) "Maggiore della Carità", Novara, Italy, from 31 December 2012 to 31 January 2014 and 47.6% of 710 infants with at least one foreign parent. 232 infants were wild-type (68.8%) for hemoglobin variants; 48 subjects (14.2%) had no hemoglobin variants, but we could not exclude the presence of a thalassemia trait (Hb A < 15%): a further monitoring of hemoglobin electrophoresis at 6 months was therefore recommended. 20 infants (5.9%) had Hb S (7.7% ± 3 of the total hemoglobin; range 3.5 - 13) and were diagnosed as Hb S carriers and 2 infants (0.6%) had Hb C (7.8% and 12.1% of the total hemoglobin, respectively) and were diagnosed as Hb C carriers. Based on our results, we can conclude that: (i) the sickle-cell disorder (Hb S) is relatively high in our territory, with a heterozygous frequency in infants at risk of 5.9%; (ii) the neonatal screening for hemoglobin disorders appears to be a valid, easy to perform test, which allows an early diagnosis and timely payment of hemoglobinopathies in populations at risk.

Neonatal Screening of Sickle Cell Anemia: A Preliminary Report

To evaluate feasibility of systematic neonatal screening for sickle cell disease in Chhattisgarh. A pilot study was done from February 2008 through January 2009 in Department of Pediatrics & Neonatology, Pt. J.N.M. Medical College & Dr.B.R.A.M. Hospital, Raipur (Chhattisgarh) on a total of 1,158 neonates. Blood samples from the neonates were taken after 48 h of birth on filter paper for detection of sickle cell anemia using Biorad hemoglobin variant Neonatal sickle cell short programme by high performance liquid chromatography (HPLC). On follow up, cases were analyzed by HPLC using Beta thalassemia short program to rule out false negative case and other hemoglobin variants. Of the 1,158 neonates screened, 628 were boys (54.2%) and 530 were girls (45.8%). Sickle cell disease was found in 3 cases (0.2%) (95%C.I 0.12-0.28), sickle cell trait was found in 68 cases (5.8%) (95%C.I 4.5-7.5). After 6-9 mo of age three cases of sickle cell diseases were reinvestigated, out of which one case turned out to be double heterozygous for sickle cell and beta thalassemia trait. Fourteen preterm neonates reported as normal in initial screening were called for follow up after 6 mo of age, 10 infants reported in OPD and 4 lost in follow up. These 10 infants were reinvestigated; 2 had sickle cell disease, 1 had sickle cell trait and 7 infants were normal. Sixty eight cases of sickle cell trait found with initial screening were also called for follow up after 6 mo of age; 61 cases reported in OPD between 6 mo to 1 y of age and 7 cases lost in follow up. Sixty one infants were reinvestigated; 60 had sickle cell trait and 1 had sickle cell disease which was reported earlier as Sickle cell trait (FAS). Thus on total follow up of cases, there were 5(0.4%) sickle cell disease, 61(5.26%) sickle cell trait, 1(0.08%) double heterozygous for sickle cell and beta thalassemia trait which needs mutation studies for thalassemia characterization (s/β(0) or s/β(+)). Early detection of sickle cell disease (SS) done by neonatal screening will help in early prevention and management of complications in postnatal period.

Molecular Analysis of Iranian Families with Sickle Cell Disease

Sickle hemoglobin is a mutant hemoglobin in which valine has been substituted for the glutamic acid normally at the sixth amino acid of the beta-globin chain. Detection of the single base pair mutation at codon 6 of the beta-globin gene is important for the prenatal diagnosis of sickle cell anemia and sickle cell disease. Application of the polymerase chain reaction technology to detect sickle cell patients and heterozygous carriers in a group of patients suspected for sickle cell disease was carried out. The sample was composed of 52 normal individuals and 52 unrelated outpatients who were attending the Hematology Research Center. All patients were interviewed. Results of their medical histories, physical examination, and the hematological analysis were recorded. The blood samples were collected in EDTA and genomic DNA was extracted from leukocytes. An amplified 110 base pair fragment of the beta-globin gene containing codon 6, was digested with the restriction enzyme MS II, and electerophoresed in 3 per cent agarose. We have established the technical condition for detection of sickle cell disease using a PCR assay. Fifteen patients having haemoglobin (Hb SS) and 37 patients in the heterozygous state (Hb AS) were identified. We confirm that the normal controls have the Hb AA genotype. The standardization of a highly sensitive and specific diagnostic test for sickle cell disease permitted us to identify the normal controls, the homozygotes and heterozygotes. This amplification method is rapid, sensitive and simple, and also has application research that is important for the prenatal diagnosis of sickle cell disease.

Compound heterozygosity Hb S/Hb Hope (beta 136 Gly-->Asp): a pitfall in the newborn screening for sickle cell disease.

The presence of Hb Hope associated with Hb S may represent a pitfall (false positive) in the neonatal detection of sickle cell disease by two of the most widely used analytical methods in screening programmes-isoelectric focusing (IEF) and high performance liquid chromatography (HPLC). This example illustrates the need to improve analytical strategies to avoid unnecessary anxiety and summoning of families often from a cultural background in which testing of the father is difficult to obtain. It is suggested that using two independent HPLC procedures might improve the specificity of the screening strategies. Additionally, simple procedures for detection of the most common mutations of the beta globin gene of DNA extracted from dried blood specimens could be easily developed for the control of abnormal samples. These procedures could be introduced into the analytical strategy.

An Audit of Sickle Cell Screening in a Paediatric Hospital Population

Current UK guidelines for the detection of sickle cell disease (SSD) recommend pre-operative screening of patients of African or Afro-Caribbean descent and of patients from the Middle East, South Italy, Greece and India. These recommendations have considerable cost implications for the NHS. We have undertaken a retrospective audit of the use of SSD screening in our own institution. Between January 1987 and August 1992, 1305 patients under 15 years of age (mean 3.8 years) underwent screening for SSD. Surgical patients accounted for 1120 tests (85.8%). The majority of patients tested were of Asian descent (n = 955; 73%); only 350 (27%) were of African or Afro-Caribbean descent. Only four patients (0.3%) tested positive (all sickle cell trait); three of Afro-Caribbean descent and one from Bahrain. We conclude that, in our population, the incidence of HbS is low. Patients of African or Afro-Caribbean descent should continue to be tested but a selective policy may be indicated in other ethnic groups.

Acute splenic sequestration crisis in sickle cell disease: Early detection and treatment

Acute splenic sequestration crisis (ASSC) in children with various forms of sickle cell disease can result in lifethreatening circulatory collapse due to the loss of circulating blood volume. Over a 6-year period we have treated 12 patients ranging in age from 5 1 2 months to 7 years presenting with acute sequestration crisis. Eleven had homozygous sickle cell disease and the other had sickle-thalassemia. One patient died of acute circulatory collapse. Eight patients underwent splenectomy after a major episode of sequestration with no serious infectious complications up to 5 years following splenectomy. Three patients with minor episodes have been followed with no recurrences. To foster early detection of this potentially lethal complication of sickle cell disease, an educational program in our Comprehensive Sickle Cell Center instructs the parents to examine the spleen and bring their child in for evaluation if the spleen enlarges. A newly developed videotape describes the common symptoms of ASSC and illustrates the technique of palpating the spleen. With early detection of sickle cell disease by neonatal screening and the educational program, the morbidity and mortality from this complication of sickle cell disease can be reduced.

Selected Testing of Newborns for Sickle Cell Disease

There are two main reasons for sickle cell testing: the early detection of those with sickle cell disease and the detection of the carrier state, sickle cell trait. The mortality of the severe forms of sickle cell disease is particularly high during the first 5 years of life. Recent data have shown that early detection of sickle cell disease and institution of expert medical care and follow-up may reduce early mortality and morbidity. The ability to detect most forms of sickle cell disease in the newborn period has made the goal of preventive care theoretically possible. The purpose of the detection of the carrier state, sickle cell trait, is less clear. Sickle cell trait has almost no clinical importance to the individual. Although a few case reports have suggested that sickle cell trait may not be a benign condition, most experts agree that sickle cell trait does not significantly alter health or disease. The main purpose of carrier testing is for genetic counseling. Mass screening of African Americans for sickle cell trait was advocated in the 1960s. However, no data have shown that results of such screening have had any impact on reproductive planning. Ideally, sickle cell testing should provide the following results. All newborns with various forms of sickle cell disease should be diagnosed by 3 months of age, before the onset of clinical problems. All heterozygotes for the βs-globin gene, or the genes for other hemoglobinopathies that may combine with βs gene to produce clinically significant forms of sickle cell disease, should be aware of the relevant genetic information at childbearing age.

Prediction and diagnosis of sickling disorders in neonates.

A pilot scheme for the prediction and detection of sickle cell disease in neonates was set up and the results from its first three years of operation analysed. A total of 153 women booking at the antenatal clinic were found to have haemoglobin S. The protocol required that all partners of the women so identified be screened for abnormal haemoglobins and beta thalassaemia trait, and that the babies of these women should have cord blood electrophoresis performed. In fact this was only achieved in 75 partners (49%) and 91 of 145 infants (63%). Of 10 babies born with sickle cell disease during the study period, nine had mothers known to have haemoglobin S or C, but only four of these mothers' partners had been tested before delivery of the child. This pilot study highlights the problems in establishing effective antenatal screening programmes, especially in ethnic minority groups. The results show that detailed staff training, careful organisation of administrative arrangements, and education of 'at risk' groups should all be undertaken before the launching of major screening initiatives.

Toward providing parents the option of avoiding the birth of the first child with Cooley's anemia: response to hemoglobinopathy screening and counseling during pregnancy.

Will the development of safe, accurate prenatal diagnosis for Cooley's anemia result in a reduction in the unwanted birth of children with Cooley's anemia in the United States? Since detection of couples at risk only through the birth of an affected child will lead to a minimal reduction in the incidence of the disease, carrier screening will be necessary. A pilot project of prenatal hemoglobinopathy screening is being conducted in Rochester, New York, to answer the question "Should hemoglobinopathy carrier screening and genetic counseling of positives be part of routine prenatal care?" In the first 27 months of this study, 6,641 unselected pregnant women have been screened on the first prenatal visit, regardless of race and ethnic origin. Of these, 304 have been positive for some type of hemoglobinopathy. Of the 293 analyzed here, the proportion coming for counseling was 61% of the total group and 77% of the thalassemia trait subgroup. Of this number, the proportion wanting their mate tested was 98% of the total group and 100% of the thalassemia trait subgroup. The proportion of those counseled whose mate was actually tested was 60% for the total group and 70% for the thalassemia trait subgroup. The proportion of at risk couples wanting amniocentesis (mostly for detection of sickle cell disease) was 61%. We conclude that, when comprehensive hemoglobinopathy screening is incorporated into routine prenatal screening, the majority of positive women make an extra visit to receive an explanation, nearly all women coming for counseling want their mate tested, the majority of mates come for testing, and the majority of couples at risk want amniocentesis. Further, pregnant women with beta-thalassemia trait may be more inclined to act upon identification as a carrier than are positive women at large.

Screening Cord Bloods for Detection of Sickle Cell Disease in Jamaica

Abstract A cord-blood screening program, designed primarily for detecting sickle cell disease, has been in operation for seven months (8000 samples) at a large maternity unit in Kingston, Jamaica. We describe techniques of cord-blood collection and electrophoretic investigation on both cellulose acetate and agar gel. These methods appear to give rapid, valid results at minimal expense and are well adapted to screening large populations.

Hospital Detection of Sickle Cell Disease

Since sickle cell disease mimics many other disorders, it may be misdiagnosed unless screening is performed routinely; in addition, its presence tends to complicate other diseases and require alterations in the way these are managed. Emphasis is placed on the avoidance of unnecessary surgery, the proper choice of anesthesia when it is necessary, and a variety of other measures that can help ward off crises.

An Automated Mass Screening Program for Sickle Cell Disease

The automated dithionite test, which identifies hemoglobin S at a reagent cost of 2 cents each and at a rate of 120 specimens per hour, has been used successfully in both civilian and military mass screening programs for the detection of sickle cell disease. Black school children in Grand Rapids, Mich, were used as the initial test population for a trail of the technique in an ordinary community. Families of those students with positive tests were selected out of a total black population for additional testing. Several patient-controlled options were then offered to afflicted families, including choice of physician, clinics, treatment, and genetic counseling. Country-wide adoption of this automated method should lead to the accumulation of accurate statistics for the first time on incidence, morbidity, mortality, gene frequency, and susceptibility to other diseases of sickle-cell-afflicted individuals.