Investigating artificial intelligence models for predicting joint pain from serum biochemistry.

OBJECTIVE: The study used machine learning models to predict the clinical outcome with various attributes or when the models chose features based on their algorithms. METHODS: Patients who presented to an orthopedic outpatient department with joint swelling or myalgia were included in the study. A proforma collected clinical information on age, gender, uric acid, C-reactive protein, and complete blood count/liver function test/renal function test parameters. Machine learning decision models (Random Forest and Gradient Boosted) were evaluated with the selected features/attributes. To categorize input data into outputs of indications of joint discomfort, multilayer perceptron and radial basis function-neural networks were used. RESULTS: The random forest decision model outperformed with 97% accuracy and minimum errors to anticipate joint pain from input attributes. For predicted classifications, the multilayer perceptron fared better with an accuracy of 98% as compared to the radial basis function. Multilayer perceptron achieved the following normalized relevance: 100% (uric acid), 10.3% (creatinine), 9.8% (AST), 5.4% (lymphocytes), and 5% (C-reactive protein) for having joint pain. Uric acid has the highest normalized relevance for predicting joint pain. CONCLUSION: The earliest artificial intelligence-based detection of joint pain will aid in the prevention of more serious orthopedic complications.

Computational imaging for rapid detection of grade-I cerebral small vessel disease (cSVD).

An early identification and subsequent management of cerebral small vessel disease (cSVD) grade 1 can delay progression into grades II and III. Machine learning algorithms have shown considerable promise in medical image interpretation automation. An experimental cross-sectional study aimed to develop an automated computer-aided diagnostic system based on AI (artificial intelligence) tools to detect grade 1-cSVD with improved accuracy. Patients with Fazekas grade 1 cSVD on Non-Contrast Magnetic Resonance Imaging (MRI) Brain of age >40 years of both genders were included. The dataset was pre-processed to be fed into a 3D convolutional neural network (CNN) model. A 3D stack with the shape (120, 128, 128, 1) containing axial slices from the brain magnetic resonance image was created. The model was created from scratch and contained four convolutional and three fully connected (FC) layers. The dataset was preprocessed by making a 3D stack, and normalizing, resizing, and completing the stack was performed. A 3D-CNN model architecture was designed to train and test preprocessed images. We achieved an accuracy of 93.12 % when 2D axial slices were used. When the 2D slices of a patient were stacked to form a 3D image, an accuracy of 85.71 % was achieved on the test set. Overall, the 3D-CNN model performed very well on the test set. The earliest and the most accurate diagnosis from computational imaging methods can help reduce the huge burden of cSVD and its associated morbidity in the form of vascular dementia.

The need for valproic acid as prophylaxis in neurosurgery patients.

Using anti-epileptic drugs in neurosurgery patients is a routine practice. This controlled trial aimed to assess whether prophylaxis with Valproate in brain surgery patients is justified or not. Group A (n=50; controls) patients received valproic acid postoperatively for three months, while group B (n=50; subjects) received a placebo. Serum valproic acid levels between 50-125g/ml were required. Kendall's Tau was applied to see the correlation between the 'frequency of seizures' between different surgical procedures performed and the extent of manipulations-EOMs. A wireless EMOTIV EPOC device was used to visualize the Electroencephalogram patterns. In controls, 12 patients had one seizure and only two patients had 2 seizures. In the placebo group, 13 patients had one and 4 patients had 2 seizures. The seizure frequency was highest amongst brain tumor patients. An insignificant difference was found between the seizure frequencies of the placebo and control groups. A statistically insignificant correlation was found between seizure frequency and independent variables: surgical procedures and EOM (%). Using an AED or not, the frequency of seizures did not substantially reduce over the postoperative period. If not necessary, the anti-epileptic medication that is frequently provided as a prophylactic against seizures in the post-operative period should not be administered.

Efficacy of azathioprine and methotrexate in patients with chronic inflammatory demyelinating polyneuropathy (CIDP).

The study examined the efficacy of various immunosuppressants in patients with chronic inflammatory demyelinating polyneuropathy. We compared the efficacy of Azathioprine and Methotrexate in the treatment of CIDP. Patients of either gender aged ≥18 years having chronic polyneuropathy progressive for at least 8 weeks having no serum para protein or any genetic abnormality and fulfilling the Koski criteria. To measure the efficacy, Overall Neuropathy Limitation Scale (ONLS) was used. Group 1 was treated with a combination of oral steroids i.e., Prednisolone and Azathioprine while group 2 was treated with a combination of Prednisolone and Methotrexate. ONLS was statistically insignificant in the patient groups (AZA versus MTX) at the beginning of the therapy (from 1-3 months) in both groups. However, in the 4th month, the AZA group performed better than the MTX group. At the 12th month, the mean ONLS score of the patients in the AZA group was 3.69, while the mean ONLS score of the patients in the MTX group was 5.30 (p-value=0.001). We concluded that Azathioprine was more efficacious as compared to Methotrexate in the treatment of CIDP based on ONLS and should be considered as a first-line immunosuppressant in the treatment of CIDP in low-income countries like Pakistan.

Death receptor pathway genes (Caspase and BID) expression analysis in cancerous cells in response to extracts of amla and turmeric.

Plant extracts antiproliferative effects were determined by using mammalian cells along the expression profile of Caspases 3, 8 and the BID gene of the death receptor-induced pathway. Two medicinal plants viz., Turmeric (Curcuma longa) and Amla (Emblica officinalis) extracts were examined for antiproliferative effect through Neutral Red-Dye uptake assay on Vero and MDA-MB 231 cell lines. A reverse transcriptase polymerase chain reaction was used to determine the expression of genes while GAPDH expression was used as an internal control. Expression of BID was up-regulated in methanolic turmeric extract-induced MDA-MB 231 cells while Caspases 3,8 expressions were the same in induced and uninduced MDA-MB 231 cells. Activated BID cleaved into tBID and activated the intrinsic pathway which caused death in methanolic turmeric extract-induced cancerous cells. Ethanolic extracts of turmeric exerted the strongest antiproliferative effects on Vero and methanolic extracts on MDA-MB 231 cells. The morphological studies of cell lines and gene expression analysis of turmeric methanolic extract-treated cells showed activation of apoptosis via converting BID into t-BID (intrinsic pathway) and activating Caspase-3 and Caspase-8 (extrinsic pathway). With the differential cytotoxicity and induction of apoptosis in induced cancer cells in comparison to uninduced cancerous cells, hence turmeric is a natural source of new anti-cancerous compounds.

Gender bias and 'doctor brides'. A social dilemma of medical students in Pakistan.

OBJECTIVE: To assess gender bias, discrimination and bullying at medical schools, and to explore the phenomenon of 'doctor brides'. METHODS: The multicentre survey was conducted at 14 medical education institutions across Pakistan from September 2020 to April 2021, and comprised medical students of either gender at both public-sector and private-sector institutions. The survey questions explored beliefs, experiences and knowledge related to common stereotypes and social issues in medical education, including female role models, work-life balance, gender roles, lack of support from family and faculty, and bullying. Association between gender with survey variables was explored. Data was analysed using SPSS 26. Thematic analysis was used to exploring knowledge around 'doctor-brides'. RESULTS: Of the 377 subjects, 245(65%) were females. The overall mean age was 21.4±1.8 years. There were 211(53.8%) subjects aged 21-23 years, and 368(97.6%) were Muslims. Significantly more women than men were of the opinion that men are encouraged and are more likely to assume leadership roles (p=0.002). More women than males agreed that household chores and work had an impact on speciality choice (p<0.001). Most sexual assault victims were women (p<0.0001), but men generally faced more bullying and hostile behaviour (p=0.014). With regard to women being forced to quit medicine after marriage/childbirth by their in-laws/husbands or change their careers from clinical medicine to preclinical teaching, 99(26.25%) subjects knew first-hand of such cases, while 238(63.12%) had no such experience to share. CONCLUSIONS: Gender bias, discriminatory behaviour and bullying were found to be widely prevalent in medical schools across Pakistan. The general perception of 'doctor brides' needs to be revisited.

Association of 25-hydroxyvitamin D with hematological profile and anthropometry in patients with glioma.

OBJECTIVE: Gliomas are immune system suppressive tumors, and the role of vitamin D is pivotal in the immune system. This study aimed to observe if there is any significant association between the serum levels of 25-hydroxyvitamin D with hematological indices and anthropometric measurements. METHODS: A total of 75 glioma patients were included, and the information was collected on gender, age group, area, socioeconomic status, intake of vitamin D and calcium in food and supplements, skin color, sunlight exposure, body mass index, and muscle strength. A nonparametric Kendall's tau-b correlation test was performed to find a correlation between 25-hydroxyvitamin D levels and blood counts, body mass index, and muscle strength. RESULTS: The majority of patients (72%) were having low lymphocytes followed by high granulocytes and high white blood cells. The majority were having low levels of both 25-hydroxyvitamin D (84%) and calcium (73%). Patients were mainly from urban areas, and the majority belonged to middle-class families having sedentary lifestyles. The majority of patients were not taking vitamin D supplements. An insufficient amount of sunlight exposure was found in most of them. The majority of the patients were although had normal weight but weak muscle strength (74.6%). An insignificant correlation was found between 25-hydroxyvitamin D levels with the hematological indices or anthropometric measurements in brain tumor patients. CONCLUSION: Vitamin D is a powerful immune modulator, and there is a great need for sufficient amounts of sunlight exposure and vitamin D-enriched diets to prevent cancer.

Association of 25-hydroxyvitamin D with hematological profile and anthropometry in patients with glioma

SUMMARY OBJECTIVE: Gliomas are immune system suppressive tumors, and the role of vitamin D is pivotal in the immune system. This study aimed to observe if there is any significant association between the serum levels of 25-hydroxyvitamin D with hematological indices and anthropometric measurements. METHODS: A total of 75 glioma patients were included, and the information was collected on gender, age group, area, socioeconomic status, intake of vitamin D and calcium in food and supplements, skin color, sunlight exposure, body mass index, and muscle strength. A nonparametric Kendall's tau-b correlation test was performed to find a correlation between 25-hydroxyvitamin D levels and blood counts, body mass index, and muscle strength. RESULTS: The majority of patients (72%) were having low lymphocytes followed by high granulocytes and high white blood cells. The majority were having low levels of both 25-hydroxyvitamin D (84%) and calcium (73%). Patients were mainly from urban areas, and the majority belonged to middle-class families having sedentary lifestyles. The majority of patients were not taking vitamin D supplements. An insufficient amount of sunlight exposure was found in most of them. The majority of the patients were although had normal weight but weak muscle strength (74.6%). An insignificant correlation was found between 25-hydroxyvitamin D levels with the hematological indices or anthropometric measurements in brain tumor patients. CONCLUSION: Vitamin D is a powerful immune modulator, and there is a great need for sufficient amounts of sunlight exposure and vitamin D-enriched diets to prevent cancer.

Machine learning models to predict neuropsychiatric disorders in various brain tumors.

OBJECTIVE: Neuropsychiatric disorders in brain tumor patients are commonly observed. It is difficult to anticipate these disorders in different types of brain tumors. The goal of the study was to see how well machine learning (ML)-based decision algorithms might predict neuropsychiatric problems in different types of brain tumors. METHODS: 145 histopathologically-confirmed primary brain tumors of both gender aged 25-65 years of age, were included for neuropsychiatric assessments. The datasets of brain tumor patients were employed for building the models. Four different decision ML classification trees/models (J48, Random Forest, Random Tree & Hoeffding Tree) with supervised learning were trained, tested, and validated on class labeled data of brain tumor patients. The models were compared in order to determine the best accurate classifier in predicting neuropsychiatric problems in various brain tumors. Following categorical attributes as independent variables (predictors) were included from the data of brain tumor patients: age, gender, depression, dementia, and brain tumor types. With the machine learning decision tree/model techniques, a multi-target classification was performed with classes of neuropsychiatric diseases that were predicted from the selected attributes. RESULTS: 86 percent of patients were depressed, and 55 percent were suffering from dementia. Anger was the most often reported neuropsychiatric condition in brain tumor patients (92.41%), followed by sleep disorders (83%), apathy (80%), and mood swings (76.55%). When compared to other tumor types, glioblastoma patients had a higher rate of depression (20%) and dementia (20.25%). The developed models Random Forest and Random Tree were found successful with an accuracy of up to 94% (10-folds) for the prediction of neuropsychiatric disorders in brain tumor patients. The multiclass target (neuropsychiatric ailments) accuracies were having good measures of precision (0.9-1.0), recall (0.9-1.0), F-measure (0.9-1.0), and ROC area (0.9-1.0) in decision models. CONCLUSION: Random Forest Trees can be used to accurately predict neuropsychiatric illnesses. Based on the model output, the ML-decision trees will aid the physician in pre-diagnosing the mental issue and deciding on the best therapeutic approach to avoid subsequent neuropsychiatric issues in brain tumor patients.

The pncA gene mutations of Mycobacterium tuberculosis in multidrug-resistant tuberculosis.

The pncA gene encodes pyrazinamidase enzyme which converts drug pyrazinamide to active form pyrazinoic acid, but mutations in this gene can prevent enzyme activity which leads to pyrazinamide resistance. The cross-sectional study was carried out during 2016-2017 for 12 months. The purpose of the study was to detect mutation at codon 12 and codon 85 in the pncA gene in local multidrug-resistant tuberculosis (MDR-TB) patients by developing a simple molecular test so that disease could be detected timely in the local population. DNA extracted from sputum-cultured samples from MDR-TB patients and subjected to semi-multiplex allele-specific PCR by using self-designed primers against the pncA gene. Among 75 samples, 53 samples were subjected to molecular analysis based on purified DNA quantity and quality. The primers produced 250 and 480 bp fragments, indicating the mutations at codon 12 (aspartate to alanine) and codon 85 (leucine to proline) respectively. MDR-TB was more common in the age group 21-40 years. Fifty-seven percent of samples (n = 30) were found positive for pncA mutations, whereas 43% of samples (n = 23) showed negative results. Thirteen percent of samples (n = 4) had mutations at codon 12 in which aspartate was converted to alanine, and they produced an amplified product of 480 bp. Eighty-seven percent of samples (n = 26) had mutations at codon 85 in which leucine was converted to proline and amplified product size was 250 bp. The mutations were simple nucleotide substitutions. The prevalence of mutations in which leucine was substituted by proline was higher than the mutations in which aspartate was substituted by alanine. A high prevalence of substitution mutation (CTG → CCG; leucine to proline) was detected in MDR-TB cases. Earlier detection of MDR-TB via an effective molecular diagnostic method can control the MDR tuberculosis spread in the population.

Post-vaccination (COVID-19) impacts in healthcare personnel.

The COVID-19 pandemic has brought attention back to its spread in medical staff. A survey-based study was conducted to combine general information related to COVID-19 exposures, acceptances, vaccines received, and side effects. The majority (62.3%) of healthcare professionals had acquired COVID-19 infection from hospital environment (51.5%) mainly who treated (64%) COVID-19 patients. 54% healthcare respondents expressed 'high acceptance' towards COVID-19 vaccines. 88% received COVID-19 vaccination. The majority of healthcare personnel received SinoPharm (65%). 82.3% did not acquire COVID-19 post-vaccination. 38% mild side effects were observed from vaccination. Following were the general side-effects: myalgia (18.2%), the feeling of sickness (16%), fever (15.6%), dizziness (7.8%), joint pain (7.4%), chills (4.8%), and flu (4.8%). Following were the common neurological side-effects reported: headache (18.2%), fatigue (16.5%), muscle pain (16%), numbness/tingling (3%), and migraine (2.6%). Nausea and diarrhoea were reported in only 3.5% of respondents. Bad taste was reported in only 3% of respondents. The 1.7% reported rash and itching. The majority of the healthcare professionals did not report significant side effects related to neurological, gastroenterological, skin and oral categories. To assess the vaccines' potential for substantial and long-term or chronic effects, more study with a larger sample size and a longer follow-up time is required.

Prediction of impacts on liver enzymes from the exposure of low-dose medical radiations through artificial intelligence algorithms.

OBJECTIVES: This study aimed to develop artificial intelligence and machine learning-based models to predict alterations in liver enzymes from the exposure of low annual average effective doses in radiology and nuclear medicine personnel of Institute of Nuclear Medicine and Oncology Hospital. METHODS: Ninety workers from the Radiology and Nuclear Medicine departments were included. A high-capacity thermoluminescent was used for annual average effective radiation dose measurements. The liver function tests were conducted for all subjects and controls. Three supervised learning models (multilayer precentron; logistic regression; and random forest) were applied and cross-validated to predict any alteration in liver enzymes. The t-test was applied to see if subjects and controls were significantly different in liver function tests. RESULTS: The annual average effective doses were in the range of 0.07-1.15 mSv. Alanine transaminase was 50% high and aspartate transaminase was 20% high in radiation workers. There existed a significant difference (p=0.0008) in Alanine-aminotransferase between radiation-exposed and radiation-unexposed workers. Random forest model achieved 90-96.6% accuracies in Alanine-aminotransferase and Aspartate-aminotransferase predictions. The second best classifier model was the Multilayer perceptron (65.5-80% accuracies). CONCLUSION: As there is a need of regular monitoring of hepatic function in radiation-exposed people, our artificial intelligence-based predicting model random forest is proved accurate in prediagnosing alterations in liver enzymes.

Clinical and biochemical indicators of disease severity and neurological findings in COVID-19: A study of King Edward Medical University (KEMU), Pakistan.

This study investigated the significance of difference between presence and absence of different neurological findings in COVID-19, in relation with the biochemistry. Various significant correlations in connection with the disease severity and clinical factors were also identified. 351 COVID-19 patients were included. Different laboratory/ clinical findings were investigated. Correlations Kendall's tau and Pearson Chi-Square were applied to find the correlations between severity and clinical findings. The Mann-Whitney Test was applied for a comparison between two types of neurological groups for each biochemistry parameter. Headache was reported in 28% and dizziness in 13% patients. The impaired smell and impaired taste were reported in 28.5% and 36.2% patients, respectively. The muscle pain was present in 39% patients. 80% patients had low lymphocytes & 70% had high neutrophils. 54.5% were found with high ALP. LDH was elevated in 73%. Severity was found significantly correlated with decreased oxygen saturation, age and raised levels of urea, creatinine and LDH. The groups (with/without CNS involvement) were statistically different in ALP, groups (with/without PNS involvement) in WBC, lymphocytes, neutrophils, ALP, urea, creatinine, CK, CKMB and LDH and groups (with/without MSK involvement) in WBC. Oxygen saturation, age, urea, creatinine and LDH are significant indicators of disease severity in COVID-19. The altered levels of different biochemistry can impact the neurological states of COVID-19 patients.

Less disease severity of COVID-19 among smokers: An indication of investigating nicotine as preventive measure.

The study was aimed to mention the epidemiology of smoking in Pakistani COVID-19 infected patients along with the disease severity, oxygen dependency and fatality rate. A cross sectional epidemiological study was carried out on 555 confirmed cases of COVID-19 infection. The median age was 47±16 years. 59% were male and 41% were female. Most of the patients (97.5%) survived, while only 2.5% expired. 25.6% patients required the oxygen. Total 17 (3%) COVID-19 patients with age 20-75 years were identified as smokers. No mortality was observed in smokers. The 1.4% smokers presented with mild disease, 1.2% with moderate disease and 0.4% had severe disease. According to Chi-Square test, there existed an insignificant difference (p-value: 0.38649) between smokers and non-smokers in disease severity levels. Smoking is a precursor for countless diseases, but it behaved differently in COVID-19 infected patients, as its prevalence was significantly low. We found no significant variation of the disease severity among the smokers and non-smokers. Profound experiments should be conducted to recommend whether nicotine can be used as a protective agent to negate COVID-19 infection.

Improvement in the quality of patient notes: A report of a closed loop audit quality improvement project at a neurosurgical department in Pakistan.

OBJECTIVE: To identify the deficiencies in patient note record-taking with the aim of improving the quality to meet international standards. METHODS: This prospective clinical quality improvement audit study was conducted at the department of Neurosurgery, Allama Iqbal Medical College, Jinnah Hospital Lahore from January 2019 to February 2020. The first audit cycle was carried out in July 2019, after data anonymisation, the notes from 1st January to 31st June were analysed in the first audit cycle against a hybrid proforma containing entries deemed essential in operative notes according to the guidelines of the Royal College of Surgeons of England. The guidelines were subsequently disseminated among postgraduate trainees using various methods. Post-intervention, randomly selected patientnotes from 1st August to 31st December 2019 were analysed in the second audit which was done in February 2020. The result of the two audits were compared to assess significance of association between the cycles for each categorical variable. RESULTS: Of the 100 patient-notes audited, 50(50%) were part of each of the two cycles. Significant improvements (p<0.05) were seen between the two cycles in time of operation, pre-op status, post-op care, monitoring instruction, mobilisation, feeding instructions, wound care and position. There was 100% improvement in entries including name, age and sex, date of operation, elective/emergency, name of the procedure and name of operating surgeon and assistant, and the name of anaesthetist. Overall, marked improvement was observed in all parameters except in 'use of antibiotic prophylaxes'. CONCLUSIONS: Regular audits are needed to monitor and improve patient care.

Effects of plants extracts on the expression of major genes of JAK/STAT pathway.

This study analyzed the effects of the plant extracts (Citrus limon, Solanum lycopersicum, Zingiber officinale, Vitis vinifera and Allium sativum) on the growth of mammalian cells (Vero and MDA-MB-231) and evaluated the most effective plant extract for the expression of specific genes of the JAK/STAT pathway in human breast cancer cells. An antiproliferative bioassay involving neutral red-dye uptake was used to determine the anticancerous potential of plant extracts. In Vero cells, the ginger methanolic extract was least effective; whereas the lemon methanolic extract was more effective with 64 dilutions with IC50 51.42%. In MDA-MB-231 cells, the tomato and ginger methanolic, and grape water extracts were least effective, whereas lemon water extract was most effective with 32 dilutions with IC50 48.67%, by upregulating JAK1, JAK2, TYK2, IRF7 and IRF3 gene expressions of the JAK/STAT pathway. C. limon inhibited the growth of both Vero and MDA-MB 231 cells. It suggested that C. limon has anti-cancer potential by inducing the JAK/STAT pathway.

Prediction of impacts on liver enzymes from the exposure of low-dose medical radiations through artificial intelligence algorithms

SUMMARY OBJECTIVES: This study aimed to develop artificial intelligence and machine learning-based models to predict alterations in liver enzymes from the exposure of low annual average effective doses in radiology and nuclear medicine personnel of Institute of Nuclear Medicine and Oncology Hospital. METHODS: Ninety workers from the Radiology and Nuclear Medicine departments were included. A high-capacity thermoluminescent was used for annual average effective radiation dose measurements. The liver function tests were conducted for all subjects and controls. Three supervised learning models (multilayer precentron; logistic regression; and random forest) were applied and cross-validated to predict any alteration in liver enzymes. The t-test was applied to see if subjects and controls were significantly different in liver function tests. RESULTS: The annual average effective doses were in the range of 0.07-1.15 mSv. Alanine transaminase was 50% high and aspartate transaminase was 20% high in radiation workers. There existed a significant difference (p=0.0008) in Alanine-aminotransferase between radiation-exposed and radiation-unexposed workers. Random forest model achieved 90-96.6% accuracies in Alanine-aminotransferase and Aspartate-aminotransferase predictions. The second best classifier model was the Multilayer perceptron (65.5-80% accuracies). CONCLUSION: As there is a need of regular monitoring of hepatic function in radiation-exposed people, our artificial intelligence-based predicting model random forest is proved accurate in prediagnosing alterations in liver enzymes.

Identification of significant risks in pediatric acute lymphoblastic leukemia (ALL) through machine learning (ML) approach.

Pediatric acute lymphoblastic leukemia (ALL) through machine learning (ML) technique was analyzed to determine the significance of clinical and phenotypic variables as well as environmental conditions that can identify the underlying causes of child ALL. Fifty pediatric patients (n = 50) included who were diagnosed with acute lymphoblastic leukemia (ALL) according to the inclusion and exclusion criteria. Clinical variables comprised of the blood biochemistry (CBC, LFTs, RFTs) results, and distribution of type of ALL, i.e., T ALL or B ALL. Phenotypic data included the age, sex of the child, and consanguinity, while environmental factors included the habitat, socioeconomic status, and access to filtered drinking water. Fifteen different features/attributes were collected for each case individually. To retrieve most useful discriminating attributes, four different supervised ML algorithms were used including classification and regression trees (CART), random forest (RM), gradient boosted machine (GM), and C5.0 decision tree algorithm. To determine the accuracy of the derived CART algorithm on future data, a ten-fold cross validation was performed on the present data set. The ALL was common in children of age below 5 years in male patients whole belonged to middle class family of rural areas. (B-ALL) was most frequent as compared with T-ALL. The consanguinity was present in 54% of cases. Low levels of platelets and hemoglobin and high levels of white blood cells were reported in child ALL patients. CART provided the best and complete fit for the entire data set yielding a 99.83% model fit accuracy, and a misclassification of 0.17% on the entire sample space, while C5.0 reported 98.6%, random forest 94.44%, and gradient boosted machine resulted in 95.61% fitting. The variable importance of each primary discriminating attribute is platelet 43%, hemoglobin 24%, white blood cells 4%, and sex of the child 4%. An overall accuracy of 87.4% was recorded for the classifier. Platelet count abnormality can be considered as a major factor in predicting pediatric ALL. The machine learning algorithms can be applied efficiently to provide details for the prognosis for better treatment outcome. Graphical Abstract Identification of significant risks in pediatric acute lymphoblastic leukemia (ALL) through machine learning (ML) approach.

Assessment and prediction of restless leg syndrome (RLS) in patients with diabetes mellitus type II through artificial intelligence (AI).

This study aimed to diagnose the incidence of restless leg syndrome (RLS) in patients with diabetes mellitus (DM) type-2, thorough artificial intelligence based multilayer perceptron (MLP). 300 cases of diabetes mellitus type-2, of age between 18-80 years were included. Point-biserial correlation/Pearson Chi-Square correlations were conducted between RLS and risk factors. We trained a backpropagation MLP via. supervised learning algorithm to predict clinical outcome for RLS. Majority of the patients were having hypertension (63%) and with peripheral neuropathy (69%). Two mostly reported scaled parameters were: 18% 'tiredness' and 14%, 'impact on mood'. A significant correlation was found in RLS with smoking, hypertension and chronic renal failure (CRF). MLP model achieved more than 95% accuracy in predicting the outcome with cross entropy error 0.5%. Following scaled symptomatic variables: 'need/urge to move' (100%) achieved the highest normalized importance, followed by 'relief by moving' (85.7%), 'sleep disturbance' (62%) and 'impact on mood' (51.3%). Artificial intelligence based models can help physicians to identify the pre diagnose RLS, so that active measures can be taken in time to avoid further complications.

Molecular detection of hupB gene of Mycobacterium tuberculosis in young patients with acute lymphoblastic leukemia (ALL) on Chemotherapy.

Cancer chemotherapy can lead to the mycobacterial infections. Tuberculosis has been reported a serious complication in leukemia patients who undergo chemotherapy. The study was focused to find mutations in hupB gene of M. tuberculosis in 50 acute lymphoblastic leukemia (ALL) patients through semi multi complex PCR. A column based DNA isolation method was adopted for DNA isolation. The gene for histone-like protein (hupB [Rv2986c]) of M tuberculosis was amplified to detect two closely related mycobacterial species. Primers M and S (histone like protein HupB) were utilized to generate amplicons of 318 bp and 291 bp for M. tuberculosis and M. bovis, respectively. Out of fifty ALL patients, 21 (42%) were females and 29 (58%) were males. The prevalence of ALL was found higher in males as compared to females. The prevalence of ALL was higher in patients of age group 5-10 years. The results of the amplification showed that, the 318 bp fragment specific for M. tuberculosis was observed in seven samples (14%), while 291 bp fragment specific for M. bovis was not observed in any sample. Children with ALL were found at higher risk for tuberculosis. A risk evaluation of tuberculosis infection must be conducted before managing chemotherapy.