Dengue out break Prediction Based on Artificial Neural Networking Model using Climatic Parameters
Article Information
Dilip Kumar Sah1, Syed Shahzadul Haque2*, Sweta Shah3, Shyam Sundar Yadav4
1Department of Nephrology, Birat Medical College Teaching Hospital Tankisinuwari, Morang, Nepal
2Department of Biochemistry, Indira Gandhi Institute of Medical Sciences, Patna, India
3Department of Obs & Gyenocology, Birat Medical College Teaching Hospital Tankisinuwari, Morang, Nepal
4Department of Internal medicine, Birat Medical College Teaching Hospital Tankisinuwari, Morang, Nepal
*Corresponding author: Syed shahzadul Haque, Department of Nephrology, Birat Medical College Teaching Hospital Tankisinuwari, Morang, Nepal
Received: 31 March 2022; Accepted: 19 April 2022; Published: 31 August 2022
Citation: Biplab Ghosh and Monika Soni. Dengue out break Prediction Based on Artificial Neural Networking Model using Climatic Parameters. Archives of Internal Medicine Research 5 (2022): 452-462.
View / Download Pdf Share at FacebookAbstract
Purpose Dengue fever is a vector-borne tropical disease radically amplifiedby 30 times in occurrence between 1960 and 2010. The upsurge is considered to be because of urbanization, population growth and climate change. Therefore, Meteorological parameters (temperature, precipitation and relative humidity) have impact on the occurrence and outbreaks of dengue fever. There are not many studies that enumerate the relationship between the dengue cases in a particular locality and the meteorological parameters.This study explores the relationship between the dengue cases and the meteorological parameters.In prevalent localities, it is essential to alleviate the outbreaks using modelling techniques for better disease control.
Materials and Methods An artificial neural network (ANN) model was developed for predicting the number of dengue cases by knowing the meteorological parameters.The model was trained with 7 years of dengue fever data of Kamrup and Lakhimpur district of Assam, India. The practicality of the model was corroborated using independent data set with satisfactory outcomes.
Result and conclusion It was apparent from the sensitivity analysis that precipitation is more sensitive to the number of dengue cases than other meteorological parameters. This model would assist dengue fever alleviation and control in the long run.
Keywords
Artificial neural network; Nonlinear models; Meteorological parameters; Dengue; Aedes mosquito.
Article Details
1. Introduction
Anemia is the most common disorder of the blood and had been proved to be highly related to the cardio-vascular diseases as well as cerebrovascular accidents [1]. However, only the subtype of sickle cell anemia was reported to be highly associated with cerebro-vascular accident (CVA) [2]. Perhaps, the most relevant studies addressing these two problems are focused on the correlation between anemia and negative outcomes after ischemic stroke [3]. The cardiovascular risk profile and overall survival has already been analyzed in the Framingham study with an interesting finding that patients with low hematocrit had mostly died of non-cardiovascular causes. The mortality due to cardiovascular events was more attributed to high hematocrit group [4]. One recent study has suggested that about 6.4 % of stroke patients have iron deficiency anemia (IDA) [5].
Iron-deficiency anemia (IDA) results in decreased total body iron causing microcytic hypochromic erythropoiesis with decreased serum iron and ferritin with an increased total iron binding capacity. Ferro et al. have reported an association of Iron-deficiency anemia with papilledema, idiopathic intracranial hypertension, venous sinus thrombosis, and ischemic stroke [6]. The exact mechanisms underlying anemia and stroke syndromes could not be comprehended but, it is believed that iron deficiency may have a strong link to ischemic stroke. Thrombocytosis is often associated with iron deficiency and microcytosis produces a reduction in the red cell deformability and could produce a hypercoagulable state. Thrombus formation in the presence of an underlying Atheros-clerotic disease could have a synergistic impact by the platelet counts and function observed in iron-deficiency anemia [7]. Anemia may also worsen regional hypoxia in areas of decreased cerebral perfusion (so-called “anemic infarction”).
Thus iron deficiency may contribute to hypercoa-gulable state by affecting blood flow patterns within the vessels because of reduced deformability and increased viscosity of microcytic red blood cells. Anemic patients need more blood flow to maintain compensation for the lack of oxygen [8]. Therefore, the increase in blood flow can cause endothelial damage, leading to platelet aggregation, causing a cascade for thrombus formation. Little information is available in published literature about the relationship between iron deficiency anemia and stroke. The purpose of this study was to determine the prevalence of iron deficiency in patients with stroke as a first step to make recommendations about iron prophylactic supplementation for this particular group of patients.
2. Material and Methods
This study was conducted after the approval from College of Physicians and Surgeons, Pakistan. The permission was granted from Jinnah Postgraduate Medical Centre, Karachi Ethical Review Committee (ERC). All the patients (n=144) were enrolled from 20th august 2016 to 20th February 2017 in medical unit I of Jinnah Postgraduate Medical Centre, Karachi which is the referral tertiary care hospital and research institute. The inclusion criteria for induction of patients was an event leading to focal neurological deficit of sudden onset lasting for more than 24 hours with evidence of hypo-density on computerized tomography (CT) scan of brain. The criteria of exclusion were hemorrhagic stroke, long-term transfusion therapy (for more than 10 years), blood transfusion within three months prior to the study, history of other critical illness such as heart failure (diagnosed on ejection fraction (EF) < 25%, chronic obstructive pulmonary disease (COPD) diagnosed on forced expiratory volume (FEV) < 70% predicted), chronic kidney disease (CKD) diagnosed on serum creatinine > 3mg/dl or malignancy.
The patients fulfilling the selection criteria of the study were already consented for written consent at the time of admission. Baseline demographic data such as age, sex, diabetes mellitus (DM), hypertension (HTN), hyperlipidemia, obesity, and history of blood trans-fusion was recorded. On the same day, five milliliters of blood were drawn from a convenient peripheral vein into plain tubes for assessment of iron deficiency anemia. A database was developed on SPSS 17. Mean and SD was calculated for age and duration of stroke (in hours). Frequency and percentages were calculated for age, sex, diabetes mellitus, hypertension, hyperli-pidemia, obesity and history of blood transfusion along with outcome variable such as frequency of iron deficiency anemia. Effect modifier were controlled through stratification of age, sex, diabetes mellitus, hypertension, hyperlipidemia, obesity, and history of blood transfusion to see effect of these on outcome variable applying chi squire test taking p value <0.05 as significant.
3. Results
One hundred and forty-four patients fulfilling the inclusion criteria were included in this study. The mean age of study population was 67.63 ± 3.103 years. The mean duration of stroke was 12.73 ± 6.95 hours. Demographics showed that majority of patients were males 100 (69.44%) and below the age of 75 years 86 (59.72%) (Table 1). Co-morbidities included diabetes mellitus in 58 (40.3%), hypertension in 72 (50%), hyperlipidemia in 30 (20.8%) and obesity in 44 (30.6%) patients. There were 30 (20.8%) patients with a history of blood transfusion. Iron deficiency anemia was recorded in 15 (10.42%) patients (Table 1). Stratification of age, gender, DM, HTN, hyperli-pidemia, obesity and history of blood transfusion is mentioned in table 2.
Variable |
Categories |
N (%) |
Age |
Mean ± SD* |
67.63 + 3.103 |
< 75 years |
86 (59.7) |
|
>75 years |
58 (40.3) |
|
Gender |
Male |
100 (69.4) |
Female |
44 (39.5) |
|
Iron deficiency |
Yes |
15 (10.4) |
No |
159 (89.5) |
|
DM* |
Yes |
58 (40.3) |
No |
86 (59.7) |
|
Hypertension |
Yes |
72 (50) |
No |
72 (50) |
|
Hyperlipidemia |
Yes |
30 (20.8) |
No |
114 (79.1) |
|
Obesity |
Yes |
44 (30.6) |
No |
104 (72.2) |
|
History of blood transfusion |
Yes |
30 (20.8) |
No |
114 (79.1) |
Table 1: Demographics and clinicopathological characteristics of patients.
Variables |
Iron deficiency anemia |
P - value |
|
Yes |
No |
||
Age |
|||
Below 75 years |
08 |
78 |
0.395 |
Above 75 years |
07 |
51 |
|
Gender |
|||
Male |
08 |
92 |
0.129 |
Female |
07 |
37 |
|
Diabetes Mellitus |
|||
Yes |
08 |
50 |
0.208 |
No |
07 |
79 |
|
Hypertension |
|||
Yes |
08 |
64 |
0.500 |
No |
07 |
65 |
|
Hyperlipidemia |
|||
Yes |
07 |
23 |
0.001 |
No |
08 |
106 |
|
Obesity |
|||
Yes |
07 |
37 |
0.129 |
No |
08 |
92 |
|
History of blood transfusion |
|||
Yes |
04 |
26 |
0.382 |
No |
11 |
103 |
Table 2: Stratification of age, gender, DM, HTN, hyperlipidemia, obesity and history of blood transfusion.
4. Discussion
South-East Asia has emerged as the region to be affected by increased burden of anemia prevalent in 32.9% of women of all age groups [9]. Trends are very much similar in developing countries like Pakistan secondary to the consumption of cereal-based diets with a low bioavailability of iron, inappropriate personal hygiene, early marriage, repeated pregna-ncies, and low literacy rate as the major identifiable factors [10, 11]. Extremes of hemoglobin either low or high may have an adverse outcome. Tanne D have reported underlying anemia in 1 out of every 5 patients presented with an acute ischemic stroke [12].
Iron-deficiency anemia (IDA), defined as a decreased total body iron, is characterized by microcytic hypo-chromic erythropoiesis with low serum iron and ferritin with an elevated total iron binding capacity [13]. Iron-deficiency anemia has been associated with papilloedema, idiopathic intracranial hypertension, venous sinus thrombosis, and ischemic stroke [14-18]. The putative mechanisms underlying anemia and stroke syndromes are not completely understood and it is believed that iron deficiency, may cause ischemic stroke by several potential mechanisms [13]. Thromb-ocytosis is often associated with iron deficiency and microcytosis produces a reduction in the red cell deformability and could produce a hypercoagulable state [19]. The platelet counts and function observed in iron-deficiency anemia could act synergistically to promote thrombus formation, especially in the setting of an underlying atherosclerotic disease [19]. Anemia may also worsen regional hypoxia in areas of decreased cerebral perfusion (so called “anemic infarction”) [13]. Our results have indicated no significant association between iron deficiency and co-morbidities such as DM, HTN, obesity or blood transfusion in the past but hyperlipidemia has found a significant association with iron deficiency anemia. This could have been attributed to the development of atherosclerotic plaque in the presence of hyper-coagulable state.
The risk factors for carotid artery thrombus formation in the absence of atherosclerosis are not well chara-cterized, although cases of carotid thrombus associ-ated with iron deficiency anemia have been reported previously [21]. Akins et al. reported 3 case reports in which young women with severe IDA and thrombo-cytosis secondary to menorrhagia developed carotid artery thrombosis [22]. Idbaih has also followed up 8 patients with spontaneous thrombosis of lesion-free carotid arteries and half of the patients with sponta-neous thrombus had IDA, mostly secondary to menorrhagia [23]. Caplan et al. reported a young woman with severe iron- and folate-deficiency anemia with a non-adherent carotid thrombus attached to a fatty streak [24]. Yarnell et al. described a patient with menorrhagia, hypochromic microcytic anemia, plate-let count of 608,000 and a stroke, whose angiogram revealed a large intraluminal mass in the carotid bifurcation [25].
Thrombus formation requires platelet activation and aggregation onto an endothelial surface with subse-quent fibrin deposition. A straightforward hypothesis is that the thrombocytosis leads to thrombus form-ation; however, the correlation between high platelet count and thrombosis is poor. Abnormal platelet activation and function are probably more important than absolute platelet count [22]. Thus, iron deficiency may contribute to hypercoagulable state by affecting blood flow patterns within the vessels because of reduced deformability and increased viscosity of microcytic red blood cells [20]. Anemic patients need more blood flow to maintain compensation for the lack of oxygen. Therefore, the increase in blood flow can cause endothelial damage, leading to platelet aggre-gation, causing a cascade for thrombus formation [20]. The presence of floating thrombus in a patient with clinical and MRI evidence of stroke represents a significant therapeutic dilemma and requires imme-diate decision about treatment. Thrombus in the internal carotid artery may resolve with medical management [26]. It is recommended to anti-coagulate patients with or without antiplatelet drugs immed-iately. Initial anticoagulation for symptomatic intral-uminal carotid artery thrombosis leads to a low rate of recurrent ischemic events, and that carotid revascu-larization if indicated can be safely performed in a delayed manner [27].
Diabetes mellitus, hypertension, hyperlipidemia and smoking are the leading causes predisposing the occurrence of ischemic stroke. Early detection and treatment have a chance to prevent atherosclerotic changes and stroke [28]. Our cohort has increased percentage of patients with already existing co-morbidities such as DM, HTN, hyperlipidemia and obesity which might have influenced the outcome in terms of ischemic stroke despite the presence of anemia in the background. Our study has certain limiting factors such as small sample size from a hospital based analysis, presence of high percentage of confounding variables such as co-morbidities. But, it has highlighted an important factor contributing adverse outcome apart from other causes contributing towards ischemic stroke. This has raised a thought about necessary analysis of blood profiling to identify iron deficiency and prompt measures in the mana-gement of acute ischemic stroke.
5. Conclusion
We have identified iron deficiency anemia in 10% of cohort. These findings suggest the possibility of an underlying Iron deficiency anemia in patients causing acute ischemic stroke supporting the need for early management. Large, population based studies are required to declare anemia as a prognosticator in ischemic stroke.
References
- Chang Y-L, Hung S-H, Ling W, et al. Association between ischemic stroke and iron-deficiency anemia: a population-based study. PloS One 8 (2013): e82952.
- Strouse JJ, Lanzkron S, Urrutia V. The epidemiology, evaluation and treatment of stroke in adults with sickle cell disease. Expert Rev Hematol 4 (2011): 597-606.
- Mehta PJ, Chapman S, Jayam-Trouth A, et al. Acute ischemic stroke secondary to iron deficiency anemia: a case report. Case Rep Neurol Med 2012 (2012): 487080.
- Batur Caglayan HZ, Nazliel B, Irkec C, et al. Iron-Deficiency Anemia Leading to Trans-ient Ischemic Attacks due to Intraluminal Carotid Artery Thrombus. Case Rep Neurol Med 2013 (2013): 813415.
- Dubyk MD, Card RT, Whiting SJ, et al. Iron deficiency anemia prevalence at first stroke or transient ischemic attack. Can J Neurol Sci J Can Sci Neurol 39 (2012): 189-195.
- Ferro JM, Canhão P. Cerebral venous sinus thrombosis: update on diagnosis and mana-gement. Curr Cardiol Rep 16 (2014): 523.
- Lee J-H, Park K-J, Chung Y-G, et al. Isolated lateral sinus thrombosis presenting as cerebellar infarction in a patient with iron deficiency anemia. J Korean Neurosurg Soc 54 (2013): 47-49.
- Cortés Buelvas A. Anemia and transfusion of red blood cells. Colomb Médica Cali Colomb 44 (2013): 236-242.
- Kassebaum NJ, Jasrasaria R, Naghavi M, et al. A systematic analysis of global anemia burden from 1990 to 2010. Blood 123 (2014): 615-624.
- Soofi S, Khan GN, Sadiq K, et al. Prevalence and possible factors associated with anaemia, and vitamin B 12 and folate deficiencies in women of reproductive age in Pakistan: analysis of national-level secondary survey data. BMJ Open 7 (2017): e018007.
- Akhtar S, Ahmed A, Ahmad A, et al. Iron status of the Pakistani population-current issues and strategies. Asia Pac J Clin Nutr 22 (2013): 340-347.
- Tanne D, Molshatzki N, Merzeliak O, et al. Anemia status, hemoglobin concentration and outcome after acute stroke: a cohort study. BMC Neurol 10 (2010): 22.
- Munot PA, De Vile C, Hemingway C, et al. Severe iron deficiency anaemia and ischaemic stroke in children. Archives of Disease in Childhood 96 (2011): 276-279.
- Saleh T, Green W. Bilateral reversible optic disc oedema associated with iron deficiency anaemia. Eye 14 (2000): 672-673.
- Kaul B, Ramanarayanan S, Mahapatra H, et al. Iron deficiency masquerading as idiopathic intracranial hypertension. British Medical Journal Case Reports 2009 (2009).
- Ogata T, Kamouchi M, Kitazono T, et al. Cerebral venous thrombosis associated with iron deficiency anemia,” Journal of Stroke and Cerebrovascular Diseases 17 (2008): 426-428.
- Nicastro N, Schnider A, Leemann B. Iron-deficiency anemia as a rare cause of cerebral venous thrombosis (2012).
- Dubyk MD, Card RT, Whiting SJ, et al. Iron deficiency anemia prevalence at first stroke or transient ischemic attack. Canadian Journal ofNeurological Sciences 39 (2012): 189-195.
- Swann L, Kendra J R. Severe iron deficiency anaemia and stroke. Clinical and Laboratory Haematology 22 (2000): 221-223.
- Franchini M, Targher G, Montagnana M, et al. Iron and thrombosis. Annals of Hema-tology 87 (2008): 167-173.
- Mehta PJ, Chapman S, Jayam-Trouth A, et al. Acute ischemic stroke secondary to iron deficiency anemia: a Case Report. Case Reports in Neurological Medicine 2012 (2012).
- Akins PT, Glenn S, Nemeth PM, et al. Carotid artery thrombus associated with severe iron-deficiency anemiaand thromb-ocytosis. Stroke 27 (1996): 1002-1005.
- Idbaih I Crassard, Vahedi K, Guichard JP, et al. Thrombotic cocktail in stroke. Neurology 64 (2005): 334.
- Caplan L, Stein R, Patel D. Intraluminal clot of the carotid artery detected radiogra-phically. Neurology 34 (1984): 1175-1181.
- Yarnell P, Earnest M, Kelly G, et al. Disappearing carotid defects. Stroke 9 (1978): 258-262.
- Keung Y-K, Owen J. Iron deficiency and thrombosis: literature review. Clinical and Applied Thrombosis/Hemostasis 10 (2004): 387-391.
- Vellimana K, Kadkhodayan Y, Rich KM, et al. Symptomatic patients with intraluminal carotid artery thrombus:outcome with a strategy of initial anticoagulation. Journal of Neurosurgery 118 (2013): 34-41.
- Djelilovic-Vranic J, Alajbegovic A, Zelija-Asimi V, et al. Predilection role diabetes mellitus and dyslipidemia in the onset of ischemic stroke. Med Arch 67 (2013): 120-123.