Browsing by Author "Akter, Fahima"

Now showing 1 - 4 of 4

3D model generation of real-objects using depth and color information
(BRAC University, 2018-04) Sakib, Alium Basir; Azam, Hasnat; Jui, Tonni Das; Akter, Fahima; Alam, Md. Ashraful
We propose and demonstrate a method of generating 3D model using depth and color information of real-object using Microsoft Kinect sensor using our proposed pixel mapping algorithm. The primary idea of the method is to manipulate RGB and depth data within arrays to plot in three dimensional frame to generate the 3D colored model of any real-object. The proposed system comprises of 3 phases. The phases include acquisition of color and depth information of real objects, extraction and processing of RGB and depth data, generating 3D model using our proposed pixel mapping algorithm. Whereas, our proposed algorithm focuses on creating several 1D arrays for depth as well as the color data meaning Red, Green, Blue pixel values separately. By mapping the depth values accordingly pixel-wise 3D black and white model can be created where the background is also included. By limiting the depth values according to the objects position the background can also be removed to construct the focused object's 3d model. Furthermore, our algorithm maps the R, G, B arrays in the black and white model to add proper colors pixel-wise .As a result, the colorful RGB 3D model can be achieved.
An enhanced CNN model for classifying skin cancer
(BRAC University, 2022-01) Haider, Kazi MD Minhajul; Dhar, Mondira; Akter, Fahima; Islam, Sadia; Shariar, Syed Ragib; Hossain, Muhammad Iqbal; Mostakim, Moin
Unrepaired deoxyribonucleic acid in skin cells causes skin cancer by generating genetic abnormalities or mutations, rising day by day. Detecting and diagnosing skin cancer in its early stages is expensive and challenging, giving superior treatment options. Given the severity of these issues, researchers have generated a set of early classification techniques for skin cancer. Skin cancer is diagnosed and segregated from melanoma by looking at the symmetry, color, size, shape, and other features of lesions. While there are various computerized approaches for classifying skin lesions, convolutional neural networks (CNNs) have been demonstrated to exceed standard practices. Moreover, CNNs are a type of deep learning that has been prominent in various fields, including medical imaging. Multiple machine learning libraries have been used in this paper. Also, we have used five pre-trained models such as Inception V3, VGG-19, VGG-16, Efficient Net B7, ResNet 50 models and presented our proposed model for skin cancer classification using the HAM10000 dataset, which is an enormous skin cancer dataset. Following that, each competent model’s image detection categorization accuracy is evaluated by comparing and assessing. This research reports a maximum accuracy of 85.25% for Inception V3 models within five pre-trained models and maximum accuracy of 90.55% for our proposed model. In terms of image detection, our experimental configuration shows that our proposed model can attain the best classification accuracy rather than the other five pretrained models. Our findings are helpful in providing a comprehensive comparison and analysis of many neural networks in the categorization of skins cancer.
Drone Detection and Tracking using Deep Convolutional Neural Networks from Real-time CCTV Footage
(Scopus, 2024) Allmamun, Md; Akter, Fahima; Talukdar, Muhammad Borhan Uddin
Drones are flying objects that may be controlled remotely or programmed to do a wide range of tasks, including aerial photography, videography, surveys, crop and animal monitoring, search and rescue missions, package delivery, and military operations. Unrestrained use, however, can pose a significant threat to safety, privacy, and security through eavesdropping, flying close to prohibited locations, interfering with public events, and delivering illicit items. Hence, real-time drone detection and tracking are indispensable and appropriate measures. This study developed real-time drone detection and tracking using the most efficient deep-learning approaches. The models were fine-tuned first to suit the required purpose and yield the desired outcome. The performance of the developed system was better than that of earlier endeavors in terms of accuracy and loss. Of the seven fined-tuned models, the Xception model constantly rendered the maximum accuracy with negligible loss. The model outperformed other state-of-the-art architectures, exhibiting an accuracy and loss of 99.18% and 3.83, respectively
Studies of Drug-drug and Drug-metal Interactions of Some Selected Antidiabetic, Antihypertensive and Lipid Lowering Drugs
(©University of Dhaka, 2022-04-13) Akter, Fahima
Drug-drug and drug-metal interactions are important area of research in drug discovery as well as pharmacodynamic actions of the drugs. Drug-drug and drug-metal complexation may introduce new molecules having new and/or better therapeutic activities in the body. In this research work, various types of complexes from antidiabetic, antihypertensive and lipid lowering drugs were investigated while interacting with each other and with the metal ions like chromium(III), lead(II), zinc(II), iron(II) and copper(II). Different types of physicochemical properties and in vivo as well as in vitro pharmacological effects of the complexes were also studied. Drug-drug interactions of three antidiabetic agents (metformin, dapagliflozin and vildagliptin) with an antihypertensive drug (olmesartan medoxomil) were performed by co-evaporated dispersion method and three complexes were synthesized as olmesartanmetformin (OM), olmesartan-dapagliflozin (OD) and olmesartan-vildagliptin (OV). Three complexes of a lipid lowering drug (rosuvastatin) with an antidiabetic drug (vildagliptin), as well as an antihypertensive drug (perindopril) were synthesized viz. perindopril-vildagliptin (PV), perindopril-rosuvastatin (PR) and rosuvastatinvildagliptin (RV). Eleven drug metal complexes were also synthesized viz.Cr-metformin, Cr-glimepiride, Cr-vildagliptin, Cr-dapagliflozin, Pb-metformin, Pb-glimepiride, Pb-vildagliptin, Pbdapagliflozin, Zn-atorvastatin, Cu-atorvastatin and Fe-atorvastatin. The co-evaporated dispersion method was used for the synthesis of both drug-drug and drug-metal complexes. TLC, HPLC, FT-IR, UV, DSC and NMR studies confirmed the synthesis of the complexes. The melting points, DSC and TGA analyses demonstrated the thermal stability as well as thermochemical properties of the synthesized complexes. The thermodynamic parameters of the interactions of BSA (bovine serum albumin) with these pure drugs and synthesized complexes were observed using fluorescence quenching method. For TLC studies of the precursor drugs and complexes, the Rf values were found to be different from each other. Although, the NMR spectra of the pure drugs (metformin, dapagliflozin, vildagliptin, glimepiride, olmesartan medoxomil, perindopril, rosuvastatin, atorvastatin) and synthesized complexes (OM, OD, OV, Cr-metformin, Crglimepiride, Cr-vildagliptin, Cr-dapagliflozin, Zn-atorvastatin, Cu-atorvastatin and Featorvastatin) were recorded. No attempt was taken for in-depth analysis of the NMR spectral data. The main objective of acquiring the 1H NMR spectra of the above mentioned drugs and complexes was to see the differences in the spectral patterns between the parent drug(s) and the corresponding synthesized complex(es). Careful analysis of the 1H NMR spectra demonstrated that differences could be seen between the spectra of the parent drug and synthesized complex. This was only to show that complexes were formed which was further supported by TLC, TGA, DSC and FTIR analyses. Melting points were found to be 221-225 oC, 80-85 oC, 150-154 oC, 175-180 oC, 126-130 oC, 156-160 oC and 163-167 oC for metformin, dapagliflozin, vildagliptin, olmesartan medoxomil, perindopril, rosuvastatin and atorvastatin, respectively. The synthesized complexes exhibited melting points at 68-72 oC, 80-85 oC, 100-105 oC, 55- 60 oC, 110-115 oC, 115-118 oC, 102-107.6 oC, 106.5-111 oC and 105.6-110 oC for OM, OD, OV, PR, RV, PV, Zn-atorvastatin, Cu-atorvastatin and Fe-atorvastatin, respectively which were also different from the precursor drugs and the complexes. The DSC thermograms of metformin, dapagliflozin, vildagliptin, olmesartan medoxomil, OD, OV, OM, Cr-metformin, Cr-glimepiride, Cr-vildagliptin, Cr-dapagliflozin, Pbmetformin, Pb-glimepiride, Pb-vildagliptin and Pb-dapagliflozin revealed the melting endotherms which were found to be different from each other. The Rt of HPLC chromatograms were found to not be identical of some parent drugs and drug complexes in the same analytical conditions. The mechanism of interactions of olmesartan, dapagliflozin, vildagliptin, metformin, and OD, OV and OM complexes with BSA were studied and found as dynamic quenches because the values of Ksv were increased with increasing temperature. The thermodynamic factors were determined from the linear plot of Van’t Hoff which indicated the spontaneous (negative value of ΔG) interaction where hydrophobic interaction was the major contributing force (positive values of ΔH and ΔS) except for olmesartan. For olmesartan-BSA, it was found that ΔH ˂0 ˂ΔS which indicated that the interaction was electrostatic force driven. The binding constants and number of binding sites were also calculated and found that one mole of the reactant (drug) and/or complex interacted with one mole of BSA. On the other hand, the interaction mechanisms of rosuvastatin, perindopril, vildagliptin, RV, PV and PR with BSA were also studied and found that perindopril, vildagliptin, PV and PR showed dynamic quenches as the values of Ksv increased along with increasing temperature. But the rosuvastatin-BSA and RV-BSA systems were developed by static quenching where, Ksv values decreased with increasing temperature. The interactions of perindopril, vildagliptin, PR, PV with BSA were mediated by enthalpy driven hydrophobic bonding (negative value of ΔG). But the rosuvastatin-BSA and RV-BSA systems were driven by the Vander Waal’s forces and H-bonds (negative value of ΔH along with ΔS˂0). The binding constants and number of binding sites were also analyzed and found that one mole of the reactant (drug) and/or complex interacted with one mole of BSA. In vivo exploration of anti-diabetic activity was done on alloxan induced mice. The study revealed that after 14 days of treatment the antidiabetic drugs e.g. metformin, dapagliflozin and vildagliptin reduced the blood sugar by 39.70%, 56.73% and 51.22%, respectively while the newly synthesized complexes e.g. OM, OD and OV reduced the blood sugar by 42.95%, 50.50% and 48.66%, respectively. Hence only OM demonstrated synergistic effect as it reduced the blood sugar level more than that exhibited by metformin. Other complexes OD and OV did not show better effect than the parent drugs dapagliflozin and vildagliptin, respectively. The OM can be demonstrated as safe because the complex revealed no damage to hepatic and nephrotic tissues. But the other complexes OD and OV produced moderate to severe dysplasia in kidney and liver tissues after 14 days of treatment. All the three complexes elevated levels of serum creatinine and uric acid than that for metformin, dapagliflozin and vildagliptin itself. The levels of serum creatinine for control, metformin, dapagliflozin, vildagliptin were found as 3.8 mg/dL, 3.38 mg/dL, 3.42 mg/dL and 3.60 mg/dL, respectively but for OM, OD and OV the concentrations were found as 4.09 mg/dL, 4.56 mg/dL and 5.95 mg/dL, respectively. Uric acid levels for control, metformin, dapagliflozin, vildagliptin, OM, OD and OV were found as 17.59 mg/dL, 10.06 mg/dL, 11.37mg/dL, 16.84 mg/dL, 12.75 mg/dL, 15.64 mg/dL and 17.81 mg/dL, respectively. The serum SGPT level for control, metformin, dapagliflozin, vildagliptin, complex OM, OD and OV treated mice were calculated as 23.85 U/L, 20.28 U/L, 21.02 U/L, 21.17 U/L, 17.35 U/L, 20.15 U/L and 27.78 U/L, respectively. Serum SGOT level in mice after treatment with drugs and drug complexes were found 21.23 U/L, 18.42 U/L, 17.24 U/L, 17.70 U/L, 15.54 U/L, 18.91 U/L and 25.67 U/L, respectively for control, metformin, dapagliflozin, vildagliptin, OM, OD and OV. Serum SGPT and SGOT levels were elevated by the treatment with OV. But OM treatment revealed reduced serum SGPT and SGOT levels than by only metformin treatment (20.28 U/L to 17.35 U/L and 18.42 U/L to 15.54 U/L, respectively). Considering all the issues the complex OM can be demonstrated as safe and promising ligand and can be suggested for further extensive studies to evaluate as therapeutic agent. In case of antidiabetic activity of four Cr-complexes viz. Cr-metformin, Cr-glimepiride, Cr-vildagliptin and Cr-dapagliflozin, they improved glucose metabolism in alloxan induced hyperglycemic mice. The treatment with the Cr-complexes significantly reduced the blood glucose level than that of the positive control group mice. Among the four Cr-drug complexes, Cr-dapagliflozin complex reduced blood glucose level significantly and it was found to be 64.20% more effective than the standard dapagliflozin. The result was followed by Cr-glimepiride by 26.72% blood glucose reduction, Cr-metformin by 23.35% reduction and Cr-vildagliptin by 7.61% reduction than the standard glimepiride, metformin and vildagliptin, respectively. But Crvildagliptin and Cr-dapagliflozin showed moderate dysplasia in hepatic tissues after 14 days of treatment. So, whether Cr-complexes can offer long-term health benefits or not is still unknown as extensive toxicological data could not be established yet. In case of Pb-antidiabetic drug complexes, they did not show significant positive effect to reduce the blood glucose level. After 14 days of treatment with metformin, glimepiride, vildagliptin, dapagliflozin, it was found that the average glucose levels of mice decreased from 31.54, 30.24, 31.50 and 30.37 to 19.02, 17.20, 19.70 and 17.60 mmol/L, respectively in mice whereas the complexes Pb-metformin, Pb-glimepiride, Pb-vildagliptin and Pb-dapagliflozin did not reduce blood glucose level considerably and blood sugar levels were found as 25.82, 29.23, 25.32 and 29.32 mmol/L, respectively. Moreover, the Pb-complexes increased serum creatinine and serum uric acid levels of mice as well as produced necrosis of the hepatic and nephrotic tissues which suggested cellular damage in liver and kidney of mice. After treatment with metformin, glimepiride, vildagliptin, dapagliflozin, the serum creatinine levels of mice were found to be 3.38, 3.96, 3.60 and 3.42 mg/dL, respectively whereas for Pbmetformin, Pb-glimepiride, Pb-vildagliptin and Pb-dapagliflozin the creatinine levels were increased to 4.57, 5.36, 5.21 and 5.24 mg/dL, respectively. The levels of uric acid of the experimental mice were elevated into 53.13 from 42.91, 57.40 from 44.83, 49.36 from 40.21 and 53.32 from 41.49 mg/dL for Pb-metformin, Pb-glimepiride, Pb vildagliptin and Pb-dapagliflozin, respectively than that of metformin, glimepiride, vildagliptin and dapagliflozin. In vivo evaluation of lipid lowering activity was done on high fat diet fed rabbits and from the experiment, it was found that all the synthesized complexes viz. complex PR, PV and RV reduced cholesterol, triglycerides, low density lipoprotein (LDL) cholesterol, very low density lipoprotein (VLDL) cholesterol, non high density lipoprotein (HDL) cholesterol levels but enhanced the high density lipoprotein (HDL) cholesterol level. The complex PR decreased the levels of serum cholesterol, triglycerides, low density lipoprotein (LDL) cholesterol, very low density lipoprotein (VLDL) cholesterol to 139.92±8.23 mg/dL, 210.1±38.34 mg/dL, 81.0±10.12 mg/dL, 30.0±2.12 mg/dL and 112.0±8.79 mg/dL, respectively while the reference drug rosuvastatin lowered the levels at 144.8±9.12 mg/dL, 280.13±40.25 mg/dL, 87.0±12.10 mg/dL, 37.0±3.23 mg/dL and 114.0±9.23 mg/dL, respectively. The complex RV also decreased cholesterol, triglycerides, low density lipoprotein (LDL) cholesterol, very low density lipoprotein (VLDL) cholesterol 172.35±10.12 mg/dL, 152.01±41.45 mg/dL, 111.0±10.11 mg/dL, 27.0±3.32 mg/dL and 137±8.32 mg/dL, respectively than the reference drug rosuvastatin did. The promising complex PV reduced the LDL and VLDL to 69.0±10.65 mg/dL and 26.0±4.13 mg/dL, respectively. But rosuvastatin as well as the three newly formed complexes PR, RV, PV increased HDL cholesterol levels to 30.34±2.01 mg/dL, 28.33±2.5 mg/dL, 34.9±2.7 mg/dL and 49.44±2.3 mg/dL, respectively than that of the control group of rabbits (12.48±2.3 mg/dL). In support of lipid lowering activity the antioxidant, thrombolytic and membrane stabilizing activities of the complexes PR, PV and RV were also evaluated in vitro. Three complexes showed better thrombolytic activity than rosuvastatin. Among the complexes, the RV demonstrated the highest thrombolytic activity (29.52±0.09%) whereas PR, PV and rosuvastatin showed 26.39±0.06%, 20.97% and 20.96±0.09% activities, respectively. The synthesized complexes PR, PV and RV displayed better antioxidant activity than the lipid lowering drug rosuvastatin. For free radical scavenging activity, the highest IC50 was produced by PV (67.71 μg/mL) among all the samples followed by RV (56.83 μg/mL), PR (54.79 μg/mL) and rosuvastatin (131.6 μg/mL). Three new complexes also were investigated for membrane stabilizing activity and showed significant effect. The rosuvastatin inhibited 39.92% hemolysis of RBCs followed by RV (35.18 %), PR (26.55 %) and PV (19.43 %) in the condition of induced by hypotonic solution.