Shams Ibne Karim
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Shams Ibne Karim has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
Farzana Yeasmin
Department of Psychiatry, Bangladesh Medical University, Dhaka, Bangladesh
1 shared publication- Galib Muhammad Shahriar Himel
School of Computer Sciences, Universiti Sains Malaysia, Penang, Malaysia
1 shared publication - Kabir Uddin Sikder
Department of Physics Jahangirnagar University Dhaka Bangladesh
1 shared publication - Kh. Abdullah Al-Aff
Bangladesh University of Health Sciences, Dhaka, Bangladesh
1 shared publication - Masudul Islam
Bangladesh University of Business and Technology, Dhaka, Bangladesh
1 shared publication Nahid Mahjabin Morshed
Department of Psychiatry, Bangladesh Medical University, Dhaka, Bangladesh
1 shared publication
Latest Publications
- Open Access
- 1 July 2026
- Published by Bangladesh Academy of Sciences in Bangabandhu Sheikh Mujib Medical University Journal
- Vol. 19 (3) , e89490
First-episode psychosis revealing a low-grade astrocytoma in an adolescent with refractory focal epilepsy: A case report
Background: Brain tumours may present first with psychiatric symptoms; in a review of 165 published cases, 10.3% had psychiatric symptoms alone. Recognition is harder in adolescents, in whom first-episode psychosis is usually attributed to primary psychiatric illness. Neuroimaging is recommended when red flags such as seizures or functional decline are present. Case description and management: A 17-year-old girl presented with six days of persecutory delusions, disorganised speech and insomnia, preceded by two to three months of irritability, fatigue and academic decline. She had focal epilepsy since the age of 10, with two to three brief seizures monthly in the preceding year despite good adherence; the last was four days before psychosis onset. No neuroimaging had ever been performed, reflecting limited access in peripheral settings. Examination, blood tests, electrocardiogram and cerebrospinal fluid parameters, including an autoimmune encephalitis panel, were normal. Contrast magnetic resonance imaging showed a 5.0 x 3.7 x 3.5 cm left temporo-parietal and capsuloganglionic lesion. Risperidone 2 mg/day, increased to 4 mg/day, was given with sodium valproate 1000 mg/day and clobazam 10 mg/day. Maximal safe resection with intraoperative language mapping followed two weeks later; integrated histopathological and molecular assessment confirmed isocitrate dehydrogenase mutant astrocytoma, WHO Classification of CNS Tumours, 5th edition grade 2, Ki-67 approximately 2%. Seizures ceased immediately after surgery and psychosis remitted gradually; she remained well at two months. Conclusion: Adolescent first-episode psychosis with refractory epilepsy and functional decline warrants urgent neuroimaging, as it may reveal an occult astrocytoma. Recovery followed the combination of resection, antiseizure optimisation and antipsychotic treatment.
- 1 July 2026
- Published by Bangladesh Academy of Sciences in Bangabandhu Sheikh Mujib Medical University Journal
- Vol. 19 (3) , e89490
Open AccessFirst-episode psychosis revealing a low-grade astrocytoma in an adolescent with refractory focal epilepsy: A case report
Background: Brain tumours may present first with psychiatric symptoms; in a review of 165 published cases, 10.3% had psychiatric symptoms alone. Recognition is harder in adolescents, in whom first-episode psychosis is usually attributed to primary psychiatric illness. Neuroimaging is recommended when red flags such as seizures or functional decline are present. Case description and management: A 17-year-old girl presented with six days of persecutory delusions, disorganised speech and insomnia, preceded by two to three months of irritability, fatigue and academic decline. She had focal epilepsy since the age of 10, with two to three brief seizures monthly in the preceding year despite good adherence; the last was four days before psychosis onset. No neuroimaging had ever been performed, reflecting limited access in peripheral settings. Examination, blood tests, electrocardiogram and cerebrospinal fluid parameters, including an autoimmune encephalitis panel, were normal. Contrast magnetic resonance imaging showed a 5.0 x 3.7 x 3.5 cm left temporo-parietal and capsuloganglionic lesion. Risperidone 2 mg/day, increased to 4 mg/day, was given with sodium valproate 1000 mg/day and clobazam 10 mg/day. Maximal safe resection with intraoperative language mapping followed two weeks later; integrated histopathological and molecular assessment confirmed isocitrate dehydrogenase mutant astrocytoma, WHO Classification of CNS Tumours, 5th edition grade 2, Ki-67 approximately 2%. Seizures ceased immediately after surgery and psychosis remitted gradually; she remained well at two months. Conclusion: Adolescent first-episode psychosis with refractory epilepsy and functional decline warrants urgent neuroimaging, as it may reveal an occult astrocytoma. Recovery followed the combination of resection, antiseizure optimisation and antipsychotic treatment.
- Top Cited PapersOpen Access
- 3 February 2024
- Published by Wiley in International Journal of Biomedical Imaging
- Vol. 2024, 1-18
Skin Cancer Segmentation and Classification Using Vision Transformer for Automatic Analysis in Dermatoscopy-Based Noninvasive Digital System
Skin cancer is a significant health concern worldwide, and early and accurate diagnosis plays a crucial role in improving patient outcomes. In recent years, deep learning models have shown remarkable success in various computer vision tasks, including image classification. In this research study, we introduce an approach for skin cancer classification using vision transformer, a state-of-the-art deep learning architecture that has demonstrated exceptional performance in diverse image analysis tasks. The study utilizes the HAM10000 dataset; a publicly available dataset comprising 10,015 skin lesion images classified into two categories: benign (6705 images) and malignant (3310 images). This dataset consists of high-resolution images captured using dermatoscopes and carefully annotated by expert dermatologists. Preprocessing techniques, such as normalization and augmentation, are applied to enhance the robustness and generalization of the model. The vision transformer architecture is adapted to the skin cancer classification task. The model leverages the self-attention mechanism to capture intricate spatial dependencies and long-range dependencies within the images, enabling it to effectively learn relevant features for accurate classification. Segment Anything Model (SAM) is employed to segment the cancerous areas from the images; achieving an IOU of 96.01% and Dice coefficient of 98.14% and then various pretrained models are used for classification using vision transformer architecture. Extensive experiments and evaluations are conducted to assess the performance of our approach. The results demonstrate the superiority of the vision transformer model over traditional deep learning architectures in skin cancer classification in general with some exceptions. Upon experimenting on six different models, ViT-Google, ViT-MAE, ViT-ResNet50, ViT-VAN, ViT-BEiT, and ViT-DiT, we found out that the ML approach achieves 96.15% accuracy using Google’s ViT patch-32 model with a low false negative ratio on the test dataset, showcasing its potential as an effective tool for aiding dermatologists in the diagnosis of skin cancer.
- 3 February 2024
- Published by Wiley in International Journal of Biomedical Imaging
- Vol. 2024, 1-18
Top Cited PapersOpen AccessSkin Cancer Segmentation and Classification Using Vision Transformer for Automatic Analysis in Dermatoscopy-Based Noninvasive Digital System
Skin cancer is a significant health concern worldwide, and early and accurate diagnosis plays a crucial role in improving patient outcomes. In recent years, deep learning models have shown remarkable success in various computer vision tasks, including image classification. In this research study, we introduce an approach for skin cancer classification using vision transformer, a state-of-the-art deep learning architecture that has demonstrated exceptional performance in diverse image analysis tasks. The study utilizes the HAM10000 dataset; a publicly available dataset comprising 10,015 skin lesion images classified into two categories: benign (6705 images) and malignant (3310 images). This dataset consists of high-resolution images captured using dermatoscopes and carefully annotated by expert dermatologists. Preprocessing techniques, such as normalization and augmentation, are applied to enhance the robustness and generalization of the model. The vision transformer architecture is adapted to the skin cancer classification task. The model leverages the self-attention mechanism to capture intricate spatial dependencies and long-range dependencies within the images, enabling it to effectively learn relevant features for accurate classification. Segment Anything Model (SAM) is employed to segment the cancerous areas from the images; achieving an IOU of 96.01% and Dice coefficient of 98.14% and then various pretrained models are used for classification using vision transformer architecture. Extensive experiments and evaluations are conducted to assess the performance of our approach. The results demonstrate the superiority of the vision transformer model over traditional deep learning architectures in skin cancer classification in general with some exceptions. Upon experimenting on six different models, ViT-Google, ViT-MAE, ViT-ResNet50, ViT-VAN, ViT-BEiT, and ViT-DiT, we found out that the ML approach achieves 96.15% accuracy using Google’s ViT patch-32 model with a low false negative ratio on the test dataset, showcasing its potential as an effective tool for aiding dermatologists in the diagnosis of skin cancer.