There is a growing need for diverse, high-quality stuttered speech data, particularly in the context of Indian languages. This paper introduces Project Boli, a multi-lingual stuttered speech dataset designed to advance scientific understanding and technology development for individuals who stutter, particularly in India. The dataset constitutes (a) anonymized metadata (gender, age, country, mother tongue) and responses to a questionnaire about how stuttering affects their daily lives, (b) captures both read speech (using the Rainbow Passage) and spontaneous speech (through image description tasks) for each participant and (c) includes detailed annotations of five stutter types: blocks, prolongations, interjections, sound repetitions and word repetitions. We present a comprehensive analysis of the dataset, including the data collection procedure, experience summarization of people who stutter, severity assessment of stuttering events and technical validation of the collected data. The dataset is released as an open access to further speech technology development.

Stuttering is a speech disorder marked by interruptions in the natural flow of speech, affecting millions globally. Accurately assessing its severity-categorized as mild, moderate, or severe is essential for guiding effective interventions and enhancing speech therapy outcomes. In this study, we address the task of stuttering severity classification using two publicly available datasets: FluencyBank Timestamped and LibriStutter. We extract diverse feature representations from Whisper, Wav2Vec2.0, and MFCCs to capture both learned and handcrafted speech characteristics. These features are evaluated using three deep neural network classifiers: CNN, ResNet50 and Transformer. Among the combinations tested, Whisper-Small embeddings paired with ResNet50 yielded the highest performance, achieving an overall accuracy of 82.1 % and an F1 score of 0.778. While, the other Whisper variants also performs fairly competitive with an accuracy of 77.7 % on Base and 76.6 % on tiny model using ResNet50. Our proposed setup establishes a strong benchmark and a potential state-of-the-art-for future research in automatic stuttering severity assessment.

Automatic analysis of colonoscopy images has been an active field of research motivated by the importance of early detection of precancerous polyps. However, detecting polyps during the live examination can be challenging due to various factors such as variation of skills and experience among the endoscopists, lack of attentiveness, and fatigue leading to a high polyp miss-rate. Therefore, there is a need for an automated system that can flag missed polyps during the examination and improve patient care. Deep learning has emerged as a promising solution to this challenge as it can assist endoscopists in detecting and classifying overlooked polyps and abnormalities in real time, improving the accuracy of diagnosis and enhancing treatment. In addition to the algorithm’s accuracy, transparency and interpretability are crucial to explaining the whys and hows of the algorithm’s prediction. Further, conclusions based on incorrect decisions may be fatal, especially in medicine. Despite these pitfalls, most algorithms are developed in private data, closed source, or proprietary software, and methods lack reproducibility. Therefore, to promote the development of efficient and transparent methods, we have organized the “Medico automatic polyp segmentation (Medico 2020)” and “MedAI: Transparency in Medical Image Segmentation (MedAI 2021)” competitions. The Medico 2020 challenge received submissions from 17 teams, while the MedAI 2021 challenge also gathered submissions from another 17 distinct teams in the following year. We present a comprehensive summary and analyze each contribution, highlight the strength of the best-performing methods, and discuss the possibility of clinical translations of such methods into the clinic. Our analysis revealed that the participants improved dice coefficient metrics from 0.8607 in 2020 to 0.8993 in 2021 despite adding diverse and challenging frames (containing irregular, smaller, sessile, or flat polyps), which are frequently missed during a routine clinical examination. For the instrument segmentation task, the best team obtained a mean Intersection over union metric of 0.9364. For the transparency task, a multi-disciplinary team, including expert gastroenterologists, accessed each submission and evaluated the team based on open-source practices, failure case analysis, ablation studies, usability and understandability of evaluations to gain a deeper understanding of the models’ credibility for clinical deployment. The best team obtained a final transparency score of 21 out of 25. Through the comprehensive analysis of the challenge, we not only highlight the advancements in polyp and surgical instrument segmentation but also encourage subjective evaluation for building more transparent and understandable AI-based colonoscopy systems. Moreover, we discuss the need for multi-center and out-of-distribution testing to address the current limitations of the methods to reduce the cancer burden and improve patient care. • We present a detailed analysis of the Medico 2020 and MedAI 2021 challenges that are aimed at advancing automated polyp and instrument segmentation in colonoscopy for early colorectal cancer diagnosis by using novel deep learning methods. • To the best of our knowledge, MedAI 2021 is the first challenge to evaluate the transparency in both GI endoscopy and colonoscopy. Through the challenge, we invited the participants to list package dependencies and architecture code (with instructions for building, compiling, and training) and share trained model weights in a standardized format. Additionally, we invited participants to include the code for model evaluation and provide repository licensing information to enable others to use the code and the trained model responsibly. Moreover, we asked the participants to explain model predictions using intermediate heatmaps, perform ablation studies, conduct a thorough failure analysis, and share their code for reproducing the results. Finally, we performed a subjective evaluation by including an expert gastroenterologist in the group and gave the final transparency score based on the usefulness and understandability of the results. Our initiative aims to promote transparency in AI research and foster the development of reliable, interpretable, and trustworthy algorithms for use in medical image segmentation. • We provide a comparative analysis of the 34 proposed methods in both challenges (3 subtasks), covering small details of each team in the form of Tables, qualitative and quantitative results (failure analysis), and an in-depth analysis of the findings. • We explore trust, safety, interpretability, transparency, and generalizability issues and provide future strategies to overcome the current limitations of developed algorithms.

Manipuri is a low-resource, Tibeto-Burman tonal language spoken mainly in Manipur, a northeastern state of India. Tone identification is crucial to speech comprehension for tonal languages, where tone defines the word’s meaning. Automatic Speech Recognition for those languages can perform better by including tonal information from a powerful tone detection system. While significant research has been conducted on tonal languages like Mandarin, Thai, Cantonese, and Vietnamese, a notable gap exists in exploring Manipuri within this context. To address this gap, this work expands our previously developed handcrafted speech corpus, ManiTo, which comprises isolated Manipuri tonal contrast word pairs to study the tones of Manipuri. This extension includes contributions from 20 native speakers. Preliminary findings have confirmed that Manipuri has two unique tones, Falling and Level. The study then conducts a comprehensive acoustic feature analysis. Two sets of features based on Pitch contours, Jitter, and Shimmer measurements are investigated to distinguish the two tones of Manipuri. Support Vector Machine, Long Short-term Memory, Random Forest, and k-Nearest Neighbors are the classifiers adopted to validate the selected feature sets. The results indicate that the second set of features consistently outperformed the first set, demonstrating higher accuracy, particularly when utilizing the Random Forest classifier, which provides valuable insights for further advancements in speech recognition technology for low-resource tonal language Manipuri.

Colonoscopy video acquisition has been tremendously increased for retrospective analysis, comprehensive inspection, and detection of polyps to diagnose colorectal cancer (CRC). However, extracting meaningful clinical information from colonoscopy videos requires an enormous amount of reviewing time, which burdens the surgeons considerably. To reduce the manual efforts, we propose a first end-to-end automated multi-stage deep learning framework to extract an adequate number of clinically significant frames, i.e., keyframes from colonoscopy videos. The proposed framework comprises multiple stages that employ different deep learning models to select keyframes, which are high-quality, non-redundant polyp frames capturing multi-views of polyps. In one of the stages of our framework, we also propose a novel multi-scale attention-based model, YcOLOn, for polyp localization, which generates ROI and prediction scores crucial for obtaining keyframes. We further designed a GUI application to navigate through different stages. Extensive evaluation in real-world scenarios involving patient-wise and cross-dataset validations shows the efficacy of the proposed approach. The framework removes 96.3% and 94.02% frames, reduces detection processing time by 38.28% and 59.99%, and increases mAP by 2% and 5% on the SUN database and the CVC-VideoClinicDB, respectively. The source code is available at https://github.com/Vanshali/KeyframeExtractionNote to Practitioners—The widespread acceptance of colonoscopy procedures as a gold standard for CRC screening is constrained by the massive amount of data recorded during the process that needs to be manually reviewed. Such manual procedures are burdensome and induce human diagnostic errors. This article suggests an automated framework to extract keyframes (important frames) from colonoscopy videos that can efficiently represent the clinically relevant information captured in the video streams. This is achieved by the automated removal of uninformative and highly correlated frames, which do not add to clinical findings. The approach ensures diversity among keyframes and provides clinicians with a multi-view of polyps for easy resection. In addition, the proposed multi-scale attention-based model improves the polyp localization performance, which further helps in refining the keyframe selection process. The comprehensive experimental results corroborate that discarding insignificant frames can enhance polyp detection and localization performance and reduce computational requirements. The study estimates 30% to 60% time saving for clinicians during video screening. In clinical practices, the proposed automated framework and our designed GUI would enable surgeons to visualize the essential data better with minimal manual interventions and assist in precise polyp resection.

Various artifacts, such as ghost colors, interlacing, and motion blur, hinder diagnosing colorectal cancer (CRC) from videos acquired during colonoscopy. The frames containing these artifacts are called uninformative frames and are present in large proportions in colonoscopy videos. To alleviate the impact of artifacts, we propose an adversarial network based framework to convert uninformative frames to clinically relevant frames. We examine the effectiveness of the proposed approach by evaluating the translated frames for polyp detection using YOLOv5. Preliminary results present improved detection performance along with elegant qualitative outcomes. We also examine the failure cases to determine the directions for future work.

Thereceived signal strength(RSS) basedWi-Fifingerprint technique is not only a cost-effective means for indoor positioning but also provides reliable positioning accuracy in the indoor settings. Thus, such positioning technique has drawn many researchers$'$attention to address its several limitations like degradedpositioning accuracydue to continuous changes in surrounding environment, highpositioning overhead,storage overheadetc. To address these issues, we propose anovel weighted fusion based efficient clustering strategy(WF-ECS) for fingerprint positioning system in this paper. Our proposed techniqueWF-ECScomputes a weighted average of the group ofreference points(RPs) having similar RSS patterns and thus, creates a more perfect match between fused positional co-ordinates and RSS patterns considered for merging to a single entry. Extensive experimentation have been carried out to evaluate and compare the performances of our proposed systemWF-ECSwith the contemporary fingerprint positioning systems including our prior work using the simulation test bed, the dataset collected from our departmental building and also thebenchmark dataset. The experimental results depict that our newly proposed techniqueWF-ECScan outperform the contemporary techniques in terms ofpositioning accuracyandpositioning overheadwhile reducing thestorage overheadin real indoor settings.

Colonoscopy video acquisition and recording have been increasingly performed for comprehensive diagnosis and retrospective analysis of colorectal cancer (CRC). Reviewing video streams helps detect and inspect polyps, the precursor to CRC. However, visualizing these streams in their raw form puts a considerable burden on clinicians as most of the frames are clinically insignificant and are not useful for pathological interpretation. For improved visualization of diagnostically significant information, we have proposed an automated framework that discards the uninformative frames from raw videos. Our approach initially extracts high-quality colonoscopy frames using a deep learning model to assist clinicians in visualizing data in a refined form. Subsequently, our work validates the effectiveness of keyframe selection by employing polyp detection models. All the evaluations are performed either patient-wise or cross-dataset to suffice the real-time requirements. Experimental results show that the keyframe extraction saves reviewing time and enhances the detection performances. The proposed approach achieves a polyp detection F1-score of 79.78% (patient-wise) and 89.22% (cross-dataset) on the SUN and CVC-VideoClinicDB databases, respectively.

Phoneme boundary analysis is an important prob-lem in the domain of speech processing to identify the bound-aries between different phonetic units. It has significance in the annotation of data sets and in recognition systems as well. This problem has been addressed by various frameworks such as Hidden Markov Modeling, Artificial Neural Networks, and Deep Learning. Even though they are effective, the task is challenging for zero or low resource languages because of the less availability of data sets. Moreover, the methods that do not rely on training processes are desirable because they can produce the required results with less amount of data. In this paper, a graph-based segmentation framework for phoneme boundary detection that uses structural properties and graph eigenvalues is proposed. The method identifies the boundary points in a single scan by using the temporal and spectral variations represented as graph eigenvalues. The method is proven to be effective on a sample data set of Indian accented English words. The proposed method with detailed analysis is presented in the paper.

The Electrocardiogram (ECG) signal is used to detect cardiac abnormalities by measuring the heart's electrical activity. ECG constitutes the fiducial points P-wave, QRS complex, and T-wave. The QRS complex is the most striking waveform that comprises Q-wave, R-peak, and S-wave. This paper presents a simple, reliable, and intuitive algorithm that meets the clinical needs for real-time R-peak detection using Fractals. The proposed method preprocesses raw ECG signal to remove powerline interference and baseline wander from noisy ECG signal, followed by area calculation using mathematical morphological operators such as erosion and dilation. These operators are implemented using dynamic programming with memoization that helps in achieving accurate results in a shorter duration. The area curve is then resampled and hard thresholded to produce R-peaks. The method achieved a Sensitivity of 95.78%, Positive Predictivity of 97.53%, and a Detection Error Rate of 8.44% on the MIT-BIH Arrhythmia Database. The proposed method is highly effective for realtime applications considering the fast and low computational complexity of fractals.

Machine Learning (ML) models have impressively performed on perceptual tasks over the past few years. However, these models remain vulnerable to adversarial attacks. An adversarial attack modifies an input by adding small perturbations to cause classifier misclassification. Although many efforts have been spent on training robust forensic models against this type of attack, the mitigation of adversarial attacks and defense against such attacks remain an active problem. We propose a defense method in forensics that mainly focuses on face adversarial attacks detection. The proposed defense method is based on the Private Fast Gradient Sign Method (P-FGSM) and Weighted Local Magnitude Pattern (WLMP) features. Face adversarial attacks are generated based on P-FGSM and extracted WLMP features are provided to Support Vector Machines (SVM) for detection of face adversarial images from the original face images. The proposed defense method is evaluated on a real-world dataset and results show that it detects the adversarial images from the original dataset with an accuracy of 98.75%.

Facial landmark points that are precisely extracted from the face images improve the performance of many applications in the domains of computer vision and graphics. Face swapping is one of such applications. With the availability of sophisticated image editing tools and the use of deep learning models, it is easy to create swapped face images or face swap attacks in images or videos even for non-professionals. Face swapping transfers a face from a source to a destination image, while preserving photo realism. It has potential applications in computer games, privacy protection, etc. However, it could also be used for fraudulent purposes. In this paper, we propose an approach to create face swap attacks and detect them from the original images. The augmented 81-facial landmark points are extracted for creating the face swap attacks. The feature descriptors Weighted Local Magnitude Patterns (WLMP) and Support Vector Machines (SVM) are utilized for the swapped face images detection. The performance of the proposed approach is demonstrated by different types of SVM classifiers on a real-world dataset. Experimental results show that the proposed system effectively does face swapping and detection with an accuracy of 95%.

Characterization of vowels in spoken English plays a significant role in designing speech processing systems. In this work, spoken English vowels are analysed to find features which can help in their characterization. The outcome of the analysis led to the proposal of a novel feature which uses tree structures for the representation of vowels. In this approach, the vowels are represented as trees with their structural properties being elements in the trees. These properties are extracted by understanding the geometrical shapes of acoustic events. To prove the features that they can distinguish between vowels, a comparison is shown by calculating the distances among the new features. The computation of distance is done by employing a tree matching algorithm. The performance of the proposed features are compared against the standard MFCC features. In the analysis, speech data of Indian native speakers was used. The analysis procedures and the results obtained are presented.

In this paper, a novel approach for phoneme boundary detection is proposed. The method uses a graph based structural analysis to understand the changes in different phonemes so that a clear boundary can be drawn between the phonemes. The features of speech segments are represented in a graph data structure and the changes in signal are observed by using a Graph Edit Distance (GED) algorithm. Finally the edit distances that are computed for the speech segments are used as criteria for the boundary detection. The proposed method with results are presented.

With the advancement of image editing tools in today's world, the manipulation of images like cropping, cloning, resizing, etc., becomes an easy proposition and on the other end, checking or determining whether an image has been manipulated or not, becomes a great challenge. Copy‐move forgery in images is the most popular tampering method in which a portion of an image is copied and pasted in some other location of the same image. The detection of copy‐move forgery has become a prominent research area. This study presents a detailed review and critical discussions with pros and cons of each of copy‐move forgery detection techniques from 2007 to 2017. This study also addresses the variation in databases, issues, challenges, future directions and references in this domain.

In this study, the problem of detecting if an image has tampered is inquired; especially, the attention has been paid to the case in which the portion of an image is copied and then pasted onto another region to create a duplication or to hide some important portion of the image. The proposed copy‐move forgery detection system is based on the scale‐invariant feature transform (SIFT) features extraction and density‐based clustering algorithm. The extracted SIFT features are matched using the generalised two nearest neighbours (2NN) procedure. Thereafter, the density‐based clustering algorithm is utilised to improve the detection results. The proposed system is tested using MICC‐F220, MICC‐F2000 and MICC‐F8multi datasets. Due to the generalised 2NN matching procedure, the proposed system is able to detect multiple forgeries present in the image. Experimental results show that the performance of the system is quite satisfactory in terms of computational time as well as detection accuracy.

The processes of the distributed system considered in this paper use loosely synchronized clocks. The paper describes a method of taking checkpoints by such processes in a truly distributed manner, that is, in the absence of a global checkpoint coordinator. The constituent processes take checkpoints according to their own clocks at predetermined checkpoint instants. Since these checkpoints are asynchronous, so to determine a global consistent set of such checkpoints there must be some sort of synchronization among them. This is achieved by adding suitable information to the existing clock synchronization messages looking at which the processes synchronize their checkpoints to form a global consistent checkpoint. Communication in this system is synchronous, so, processes may be blocked for communication at the checkpointing instants. The blocked processes save the state in which they were just before being blocked. It is shown here that the set of such i-th checkpoints is consistent and hence the rollback required by the system in case failure occurs is only up to the last saved state.

Linear Quadtree (LQ), Interpolation-Based Binary tree (IBB) and Minimized Boolean Function (MBF) are some of the set-of-codes type of encoding scheme for region representation of binary images. Chain Code (CC) is a well-known technique for boundary representation of such images. It is desirable to develop efficient methods of conversion between the two types of representation schemes. In this paper, we present a generalized algorithm for conversion from set-of-codes type of representation to chain codes. The computational complexity of the algorithm is also discussed.