Yash Srivastav

Human-Robot Interaction (HRI) has become an essential area that combines AI, robotics, affective computing, and healthcare technologies. This study considers the application of artificial intelligence (AI) in managing emotional connections, multisensory interaction, and psychological support in robot-led pregnancy systems. This article focuses on emotional recognition systems, multisensory communication, biosensors, adaptive robotics, ethical questions, social acceptance issues, and future advancements in maternal healthcare robotics. There is existing evidence suggesting that the use of emotionally intelligent robot systems may help improve maternal mental state, minimize the negative effects of stress and anxiety, increase engagement in healthcare processes, and give individualized assistance during pregnancy. Machine learning (ML) techniques, natural language processing, affective computing, and physiological sensors play a major role in enhancing the emotional intelligence of robotic healthcare technologies. At the same time, there are some difficulties related to the authenticity of emotions, privacy issues, emotional dependence, biased algorithms, and other ethical concerns that restrict their application. Overall, robot-led pregnancy systems demonstrate great potential as maternal healthcare technology solutions.

Brain-Computer Interfaces (BCIs) have been proposed as assistive technologies for Locked-In Syndrome (LIS) patients that can facilitate communication based on decoding of neural signals. Traditional BCI systems based on recurrent neural network (RNN) models exhibit certain constraints in terms of decoding accuracy, communication speed, and response latency. The current study aims to assess the effectiveness of transformer-based frameworks in optimizing the efficiency of both invasive and non-invasive BCI systems as compared to classical RNN models. A computational-clinical study design was used which involved participation of 48 LIS or severely paralysed participants. Subjects were grouped in accordance with their involvement in invasive or non-invasive BCI groups, and assessments were conducted during a period of eight weeks of intervention. Neural activity data processing was done with the help of two different approaches, including transformer-based model application and RNN application, assessing communication speed, decoding accuracy, latency, and error rates of both systems. Results suggest that transformer-based neural decoding frameworks proved to be superior to RNNs in terms of all evaluated criteria. Invasive transformer-based BCI demonstrated the best results concerning communication speed, decoding accuracy, lowest latency, and lowest error rates. Non-invasive transformer BCIs also yielded better results than RNN-based BCIs.

Sudden stillbirth still poses as one of the key challenges in maternal and fetus care, especially in developing nations where sophisticated labor ward monitoring systems cannot be afforded. It becomes very challenging to detect a pregnancy at risk early due to the combination of several risk factors related to both mother and the fetus. This paper presents the design of a Human-in-the-Loop Explainable Artificial Intelligence (XA)I-based predictive labor ward model to help detect composite risks related to sudden stillbirth. For this, the research considers clinical records on 90 pregnant mothers and then utilizes machine learning (ML) models such as Logistic Regression, Random Forest, and XGBoost for predictions. XAI algorithms are utilized to enhance transparency, interpretability, and clinician understanding of predictive results. It is found that the highest prediction accuracy can be achieved by usinsg the XGBoost-XAI method, which is superior to traditional approaches. Hypertension in mother, fetal distress, placental inefficiency, gestational diabetes, and prolonged labor are some of the most significant predictors of sudden stillbirth. The Human-in-the-Loop concept makes it more reliable.

Eyelid twitching and involuntary facial muscle spasms have become common neuromuscular disorders due to stress, anxiety, sleeplessness, prolonged computer usage, exhaustion, and other external factors. The purpose of this review is to discuss various neurophysiological, psychosomatic, environmental, and medical aspects of eye twitching disorders such as eyelid myokymia, hemifacial spasm, and blepharospasm in highly stressed people. Human research demonstrates that chronic stress along with dysfunction in the autonomic nervous system plays an important role in neuromuscular hyperactivity and ocular muscle spasms. Magnesium is discussed in this review as an important nutrient for nerve signaling, muscle relaxation, and neurotransmitter function. Therefore, magnesium intake in combination with stress management methods like meditation, yoga, and sleep may help alleviate the symptoms of eyelid twitches. Neurological complications like hemifacial spasm and blepharospasm generally require the intervention of drugs, neurological procedures like botulinum toxin injection therapy, anticonvulsants, and microvascular decompression surgery. The review also touches upon the effects of prolonged muscular spasm within the eye muscles on emotions, occupation, and quality of life from a psychosocial perspective. While previous human-based studies have shed light on various clinical aspects of the subject, there remain certain issues like small sample size, variation in therapeutic protocols, and absence of longitudinal studies that underscore the need for further clinical research.

Liquid biopsy using circulating tumor DNA (ctDNA) is gaining momentum as a powerful non-invasive tool for multi-cancer early detection (MCED) and precision medicine. The current paper highlights the role of artificial intelligence (AI), encompassing machine learning and deep learning techniques, in refining the analytical process of ctDNA to facilitate early cancer detection, diagnosis, prediction of the tissue of origin, and individualized disease management. Clinical trials in humans for various cancers, such as lung, colorectal, pancreatic, breast, and ovarian cancer, illustrate the ability of AI-enhanced ctDNA technology to detect even minute molecular changes in terms of mutations, epigenetic patterns, fragmentation features, and chromosome anomalies with higher sensitivity and specificity. In addition, the review highlights the biological relevance of ctDNA, the clinical utility of AI-based MCED systems, and the benefits of non-invasive testing, continuous surveillance, and detection of multiple cancers through a single blood sample. However, significant drawbacks, including low ctDNA concentration in early-stage tumors, false positives and negatives, non-standardization, ethical issues, and expensive technology, are substantial impediments to clinical adoption. Nonetheless, AI-powered ctDNA diagnostics hold immense promise for revolutionizing cancer screening in the future.

The fusion of graphene neural interfaces with adaptive artificial intelligence (AI) systems has become a critical breakthrough in human-centred human-centred neurotechnologies and personalised healthcare. Graphene has an outstanding electrical conductivity, flexibility, transparency, light architecture and biocompatibility, making it an ideal material for wearable and implantable neural devices. At the same time, artificial intelligence systems that adapt their performance benefit the interpretation of the neural signals, learning in real time, signal recognition, and performance of rehabilitation. This review covers the structural and functional characteristics of graphene neural interfaces, adaptive AI in neural signal processing, and the synergy and application of both to brain–computer interfaces (BCIs), neuroprosthetics, assistive communication systems, and personalized neurotherapy. Humans studies show that graphene-AI systems have boosted the stability of neural signals, motor control, speech decoding and rehabilitation efficiency, as well as neural monitoring and remote healthcare. The review also covers critical issues like long-term biocompatibility, privacy of neural data, algorithmic transparency, cybersecurity, and regulatory approval. While small-scale clinical trials and the absence of standardized frameworks pose challenges, the potential applications of graphene-AI combination in neurological rehabilitation and intelligent healthcare systems are promising.

Biomimetic robotics is a cross-disciplinary area combining human biomechanics, robotics, artificial intelligence, and materials science in the development of robotic systems that can mimic human movements and functionality. The current review explores the connection between human musculoskeletal kinematics and advanced high-torque robotic actuation systems through the discussion of biomechanical theory, robotic actuator technologies, biomimetic mapping techniques, and innovative developments in the field of robotic engineering. Modern approaches such as motion capture, electromyography (EMG), inverse dynamics, biomechanical modeling, and artificial intelligence-controlled systems enable enhanced accuracy of movements, sensor fusion, and human-robot interaction in the robotic systems. Research shows that biomimetic robotics enables enhanced adaptability, efficiency, and safety during interactions with humans; nevertheless, it is still difficult to mimic the complex functionality and energy efficiency of the human musculoskeletal system. In addition, this study focuses on the innovative advancements in the field such as brain-computer interfacing and AI-enabled adaptive robotic systems.

Fetus in fetu (FIF) is a rare congenital anomaly that occurs when a malformed parasitic twin grows within the host twin (most frequently in the retroperitoneum). FIF is a rare condition, with a high diagnostic and surgical challenge for neonate and infant patients due to its similarity to retroperitoneal teratoma. The purpose of this study was to review the clinical presentation, radiological findings, surgical management and outcomes of FIF in neonates and infants by analyzing 80 cases of FIF reported between 2000 and 2025. Pediatric surgery journals, radiology reports, medical databases such as PubMed, Scopus, and Google Scholar, were used to collect data. The results indicated that the abdominal distention and palpable abdominal mass were the most common symptoms and male infants were more frequently affected. CT scan and MRI were very helpful for the identification of vertebral columns, limb buds and calcified skeletal structures, which aided in differentiating FIF from retroperitoneal teratoma. Surgical resection led to good postoperative results, with low recurrence and postoperative complications. All cases were diagnosed by histopathological examination. Prompt diagnosis and surgical intervention are still vital for favorable management and neonatal outcomes.

Infertility is becoming an increasingly common reproductive health condition globally, leading to a dramatic increase in the use of assisted reproductive technologies, including intracytoplasmic sperm injection (ICSI) and in vitro fertilisation (IVF). Numerous factors, such as the mother, the embryo, and the IVF procedure, contribute to the success rate of in vitro fertilisation (IVF) and live births. Investigated here are in vitro fertilisation (IVF) success rates as a function of oocyte quality, maternal body mass index (BMI), embryo transfer methods, and ICSI. Female infertility patients undergoing in vitro fertilisation procedures at assisted reproduction centres were the subjects of the study, which used a quantitative methodology. Embryonic factors were considered alongside age, BMI, oocyte shape, fertilisation, embryo growth, embryo implantation rate, and pregnancy success rates. A chi-square test, descriptive statistics, regression models, and correlation analyses were all used to analyse the data statistically. The results show that the mother's oocyte quality and body mass index (BMI) significantly affect live birth rates, embryo growth, embryo implantation rate, and fertilisation success. There was a correlation between poor oocyte quality and high maternal BMI, lower rates of IVF success, and lower chances of live births.

This current study focuses on the impact of Artificial Intelligence (AI) on the rapid identification of new molecules to suppress aging processes, which increases the proportion of healthspan relative to lifespan. The research approach taken involved a quantitative method, where artificial intelligence-based machine learning, bioinformatics, and statistical analysis were used alongside computational molecular docking. Biochemical information from databases such as PubChem, DrugBank, and ChEMBL was leveraged to screen and analyze molecular data. It was observed that AI-assisted predictive models, especially Deep Learning Neural Networks, offered highly accurate predictions concerning the biological activity of anti-aging compounds. The selected molecules showed considerable decreases in oxidative stress, inflammation, and cellular senescence markers, coupled with improved mitochondrial function and cell repair. Moreover, quantitative results showed that the use of AI for predicting the efficacy of anti-aging agents led to more significant healthspan enhancements than lifespan increases.