Prince Ebenezer Adjei

Mr. Prince Ebenezer Adjei

University of Electronic Science and Technology of China Chengdu, CN
+1
orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
24 November 2021 in Applied Sciences
Verified
Journal Article
Open access
PDF file
ZLZenebe Markos Lonseko
Zenebe Markos Lonseko
PAPrince Ebenezer Adjei
Prince Ebenezer Adjei
WDWenju Du
Wenju Du
CLChengsi Luo
Chengsi Luo
DHDingcan Hu
Dingcan Hu
NRNini Rao
Nini Rao
Gastrointestinal (GI) diseases constitute a leading problem in the human digestive system. Consequently, several studies have explored automatic classification of GI diseases as a means of minimizing the burden on clinicians and improving patient outcomes, for both diagnostic and treatment purposes. The challenge in using deep learning-based (DL) approaches, specifically a convolutional neural network (CNN), is that spatial information is not fully utilized due to the inherent mechanism of CNNs. This paper proposes the application of spatial factors in improving classification performance. Specifically, we propose a deep CNN-based spatial attention mechanism for the classification of GI diseases, implemented with encoder–decoder layers. To overcome the data imbalance problem, we adapt data-augmentation techniques. A total of 12,147 multi-sited, multi-diseased GI images, drawn from publicly available and private sources, were used to validate the proposed approach. Furthermore, a five-fold cross-validation approach was adopted to minimize inconsistencies in intra- and inter-class variability and to ensure that results were robustly assessed. Our results, compared with other state-of-the-art models in terms of mean accuracy (ResNet50 = 90.28, GoogLeNet = 91.38, DenseNets = 91.60, and baseline = 92.84), demonstrated better outcomes (Precision = 92.8, Recall = 92.7, F1-score = 92.8, and Accuracy = 93.19). We also implemented t-distributed stochastic neighbor embedding (t–SNE) and confusion matrix analysis techniques for better visualization and performance validation. Overall, the results showed that the attention mechanism improved the automatic classification of multi-sited GI disease images. We validated clinical tests based on the proposed method by overcoming previous limitations, with the goal of improving automatic classification accuracy in future work.

20 March 2025 in Intelligent Systems with Applications
Verified
Journal Article
Open access
ADAlbert Dede
Albert Dede
Henry Nunoo-MensahHenry Nunoo-Mensah
Henry Nunoo-Mensah
Eric TchaoEric Tchao
Eric Tchao
AAAndrew Selasi Agbemenu
Andrew Selasi Agbemenu
PAPrince Ebenezer Adjei
Prince Ebenezer Adjei
JKJerry John Kponyo
Jerry John Kponyo

18 February 2025 in Engineering Reports
Verified
Journal Article
Open access
ADAlbert Dede
Albert Dede
Henry Nunoo-MensahHenry Nunoo-Mensah
Henry Nunoo-Mensah
EAEmmanuel Kofi Akowuah
Emmanuel Kofi Akowuah
KBKwame Osei Boateng
Kwame Osei Boateng
PAPrince Ebenezer Adjei
Prince Ebenezer Adjei
JKJerry John Kponyo
Jerry John Kponyo
Convolutional neural networks (CNNs) typically compress high‐resolution images to minimize computational requirements. However, this can lead to loss of information and reduced accuracy in classification tasks. This paper introduces WaveNet, a novel approach for processing high‐resolution images using wavelet‐domain inputs in CNNs. We address the challenge of maintaining classification accuracy with high‐resolution inputs while minimizing computational complexity. Our method employs wavelet packet transform (WPT) for image pre‐processing, extracting detailed multi‐scale and directional information from high‐resolution images. We propose a wavelet‐adaptive efficient channel attention (WAECA) module to dynamically select the most informative wavelet subbands. Popular CNN architectures like ResNet‐50 and MobileNetV2 are adapted by replacing their initial convolutional layers with wavelet‐transformed inputs, enabling direct learning in the wavelet domain. Experiments conducted on the Caltech‐256 and ALOT datasets demonstrate that WaveNet improves classification accuracy while reducing computational complexity. For instance, our wavelet‐enhanced ResNet‐50 achieves a Top‐1 accuracy of 72.47% on Caltech‐256, outperforming the baseline (70.65%) while reducing FLOPs from 16.52G to 3.98G. Similar improvements are observed across different architectures and datasets. We also evaluate various wavelet filters and ResNet backbones, finding that the bior1.1 filter and ResNet‐50 provide optimal performance. This work presents a practical solution for developing more accurate and efficient models for high‐resolution inputs without extensive computational resources or complex architectural modifications.

Verified
Journal Article
Open access
PDF file
ZLZenebe Markos Lonseko
Zenebe Markos Lonseko
WDWenju Du
Wenju Du
PAPrince Ebenezer Adjei
Prince Ebenezer Adjei
CLChengsi Luo
Chengsi Luo
DHDingcan Hu
Dingcan Hu
NRNini Rao
Nini Rao
Background: Accurate gastrointestinal (GI) lesion segmentation is crucial in diagnosing digestive tract diseases. An automatic lesion segmentation in endoscopic images is vital to relieving physicians’ burden and improving the survival rate of patients. However, pixel-wise annotations are highly intensive, especially in clinical settings, while numerous unlabeled image datasets could be available, although the significant results of deep learning approaches in several tasks heavily depend on large labeled datasets. Limited labeled data also hinder trained models’ generalizability under fully supervised learning for computer-aided diagnosis (CAD) systems. Methods: This work proposes a generative adversarial learning-based semi-supervised segmentation framework for GI lesion diagnosis in endoscopic images to tackle the challenge of limited annotations. The proposed approach leverages limited annotated and large unlabeled datasets in the training networks. We extensively tested the proposed method on 4880 endoscopic images. Results: Compared with current related works, the proposed method validates better results (Dice similarity coefficient = 89.42 ± 3.92, Intersection over union = 80.04 ± 5.75, Precision = 91.72 ± 4.05, Recall = 90.11 ± 5.64, and Hausdorff distance = 23.28 ± 14.36) on the challenging multi-sited datasets, confirming the effectiveness of the proposed framework. Conclusion: We explore a semi-supervised lesion segmentation method to employ the full use of multiple unlabeled endoscopic images to improve lesion segmentation accuracy. Experimental results confirmed the potential of our method and outperformed promising results compared with the current related works. The proposed CAD system can minimize diagnostic errors.

17 May 2024
Verified
Conference Paper
Open access
PAPrince Ebenezer Adjei
Prince Ebenezer Adjei
OOObed Kojo Otoo
Obed Kojo Otoo
JYJinlang Yang
Jinlang Yang
NRNini Rao
Nini Rao
Colorectal cancer (CRC) management, particularly for invasive histopathology analysis, has benefited from the prognostic insights the tumor-stroma ratio (TSR) provides. Despite advances in colonography, histopathology, and radiomics, the relationship between TSR and radiomic features still needs to be studied, given the potential non-invasive nature of radiomics biomarkers. This study explores the relationship between TSR and radiomics that can reduce the reliance on invasive biopsies in CRC management by identifying noninvasive radiomic-based markers. A retrospective analysis was conducted on radiology slides and corresponding histopathology slides from the colorectal cancer patients (n = 15) sourced from The Cancer Genomic Atlas (TCGA). A specialist radiologist manually segmented the tumors, and radiomic feature extraction was performed on radiology slides. A pre-trained vision transformer model characterized pathology slides. Using a tile-based approach, each tile was classified into one of nine distinct classes, resulting in a complete characterization of the entire slide. TSR was estimated as the stroma ratio to tumor area on pathology slides. Statistical correlation analysis and machine learning algorithms were employed to find associations between TSR and radiomic features, including feature importance assessment. Our findings indicate a significant correlation between specific gray-level measurement features and TSR, suggesting the potential of these radiomic features as noninvasive prognostic indicators. In summary, the interaction between radiomics and tumor-stroma ratio (TSR) presents potential opportunities for improving colorectal cancer (CRC) prognosis, treatment decision-making, and patient stratification.

1 January 2026 in International Journal of Biomaterials
Verified
Journal Article
Open access
EAEyram Ahadzi
Eyram Ahadzi
OMOtema Darfour Manu
Otema Darfour Manu
RARUTH ANTWIBAAH
RUTH ANTWIBAAH
HAHenry Agbe
Henry Agbe
PAPrince Ebenezer Adjei
Prince Ebenezer Adjei
Smart hydrogels are stimuli‐responsive biomaterials that enable targeted, sustained, and controlled therapeutic delivery. Their unique ability to respond to physiological cues such as pH, temperature, reactive oxygen species (ROS), and enzymatic activity makes them highly attractive for myocardial infarction (MI) therapy. This scoping review explores advances, applications, challenges, and future perspectives related to the use of smart hydrogels in MI. An initial comprehensive search of PubMed, Scopus, and ScienceDirect yielded 1802 records, from which 32 studies published between 2011 and 2025 met the inclusion criteria under the PRISMA‐ScR framework. An updated search identified seven additional eligible studies, bringing the total to 39 included studies. These studies primarily focused on smart hydrogels for delivering bioactive agents, including growth factors, stem cells, exosomes, and nanoparticles, to enhance myocardial regeneration and repair. Evidence from preclinical models consistently demonstrated improvements in myocardial remodeling, angiogenesis, cell survival, and functional cardiac recovery. Despite these encouraging findings, several important limitations were identified. Studies relied on short evaluation periods, lacked long‐term follow‐up, and were restricted to small‐animal models, thereby limiting the robustness of translational insights. Large‐animal studies and early clinical investigations remain scarce, highlighting the gap between experimental progress and clinical application. Nonetheless, smart hydrogels stand out as a promising therapeutic platform because of their capacity for localized, adaptive, and sustained responses within the complex postinfarct microenvironment. Future work should prioritize long‐term efficacy studies, development of minimally invasive delivery methods, and the use of clinically relevant preclinical models to better facilitate translation to human MI therapy.