http://hdl.handle.net/123456789/17504 2026-07-16T20:42:13Z http://hdl.handle.net/123456789/20611 Wuqla AI 03-134221-018, Muhammad Ahsan Wuqla AI is a smart online platform created to enhance the process of accessing information regarding the property-related legal matters in Pakistan via incorporation of Chat bot. The system enables to simplify otherwise complicated legal notions like the ownership issues, inheritance question, and the process of the property transfer to make them more comprehensible and accessible to the broad audience. It offers a well-organized space where one can seek automated legal advice, find legal practitioners, launch live interaction and manage his profile effectively using just one platform. Some of the essential features of the system are AI-based legal query processing, search through lawyers by name, intelligent lawyer recommendation to suit the specific legal issue, chatting consultation, request chat, and feedback and rating system to evaluate the service. All these characteristics contribute to the convenience of users, convenience in decision-making, as they enable communication of the user with the right legal specialist and legal information that can be trusted, in an efficient way. Ms. Summaira Nosheen 2026-01-01T00:00:00Z http://hdl.handle.net/123456789/20958 DoseLogix: Pharmaceutical Distribution Management System 03-134221-046 YASIR JAMIL, 03-134221-008 ALI RAZA The main objective of this venture is to overcome major constraints of the current Pharmaceutical Distribution Management Systems especially among the small and medium sized distributors. The existing software tends to be either old-fashioned in UI/UX, inconsistent data flow, expensive, or crippling with a deficit of certain industry-specific functional support. The key aim of DoseLogix is to develop a low-cost, contemporary, and powerful PDMS tool that can easily manage such vital functions as full inventory and detailed financial monitoring. This dedicated system is specifically designed to offer a secure authentication, the ability to support multi-tenancy, and built-in advanced analytics to support robust business expansion in the future. The system architecture is a service-oriented system design with a modern Next.js front end and an asynchronous backend based on Express.js on the Node.js platform, and a dedicated Python micro-service to perform advanced computational functions. It uses JavaScript and TypeScript to develop the core applications, and uses Python libraries, including Scikit-learn and Pandas, to carry out the most important machine learning operations. The system uses MongoDB Atlas, a NoSQL document database that is a cloud-based system, to support data persistence across all the system modules with a high level of scalability and real-time data management of data. The implementation successfully provided a powerful centralized application that included lean sales and purchase invoicing, real-time inventory controls with First Expiry First Out (FEFO) logic, and delivery logs. One of its major inventions is the inbuilt AI-based module, which makes correct predictions of up to seven days of future product sales based on the historical data interpreted by the best-performing models on our dataset. The system also includes a critical web scraping system to automatically gather the most recent health advices from recognized Pakistani health advisories to help the distributors make proactive buying choices. The most important operational improvements are tracking the aging of receivables, automated stock notifications, and showing the previous three sales prices right in the sales invoice creating process. As a concluding remark, DoseLogix is able to meet its major mandate, that is, providing an efficient, secure and modernized Pharmaceutical Distribution Management System, which is specifically designed to suit small and medium-sized enterprises. This is a specialized platform that directly addresses and overcomes key industry gaps posed by old, generalized or too expensive distribution software. The effective integration of advanced, user-centric features, including AI sales forecasting, latest health news from advisories and real-time financial tracking, significantly increases operational accuracy and supply chain responsiveness. DoseLogix is positioned as a highly reliable, scalable and industry-specific software that makes life easier and promotes sustainable growth of business among pharmaceutical vendors. Dr Junaid Nasir Qureshi 2026-01-01T00:00:00Z http://hdl.handle.net/123456789/20959 Fluenti: AI Powered Speech Therapy & Emotional Support 03-134221-039 SYEDA HIRA MEHBOOB, 03-134221-034 SAMAHA MUNIR The barriers to speech therapy and emotional support also include increased accessibility and inconsistent quality, relating to both children with speech disorders and adults who need mental health assistance. Such issues can be tackled through this project the creation of Fluenti, which is the union of interactive story building games that can be used in speech therapy with AI oriented emotional support sessions aimed at adults. The architecture uses a multi layered design where the frontend will be based on React and TypeScript and the backend will be based on Node.js and Express.js, with the specialized therapist services being based on Python Flask. In the case of speech therapy, Google Gemini creates age specific interactive narratives and carries out evaluations in four types of therapies, which are pronunciation, fluency, Developmental Language Disorder (DLD), and social communication. The story game uses Web Speech APIs browser native and speech recognition in real-time, dynamically changing difficulty level, and progress tracking, which is gamified. To offer emotional support, a Python-based therapy bot, which is driven by Groq LLM, offers responses with hybrid crisis detection (AI and pattern-based) and psychological profiling and session management. The site combines OpenAI Whisper API with speech-to-text and OpenAI TTS with text-to-speech to allow the voice responses. MongoDB takes care of user progress, session history, and psychological profiles data persistence. The security features are JWT authentication, two-factor authentication, encryption and rate limiting. The system is a connection between therapeutic interventions and modern technology, which offers accessible, scaled, and individual support. The further development will focus on a broader range of types of therapies, improved crisis-detecting algorithms, and the system modification to multilingual support and global markets. Dr Junaid Nasir Qureshi 2026-01-01T00:00:00Z http://hdl.handle.net/123456789/20612 SkinCare:An AI Guide To Skin Wellness 03-134221-036, Subaktgeen Haider Skin diseases represent a significant health concern at a worldwide level, especially in areas where accessibility to dermatologists is restricted, the cost of consulting with them is high and there is a delay in seeking medical attention, which in most instances causes late diagnoses and avoidable complications. Due to the swift growth of smartphones and artificial intelligence (AI), increased interest in high-speed, reliable, and privacy-oriented tools exists that can be used to initiate early awareness of skin-health. The proposed project presents the SkinCare: An AI Guide to Skin Wellness, a mobile app that will analyze skin-condition images in real-time using a quantized TensorFlow Lite model and provides intelligent guidance to users by training a lightweight conversational assistant named Gemini Flash-Lite Pro that will assist users in taking high-quality images, interpreting AI outputs, and offering them safe educational advice. All processing is done on the device to guarantee high privacy and low latency with an average of 0.8 seconds inference time on mid-range Android machines. The HAM10000, ISIC Archive, and Fitzpatrick17k datasets were used to train the AI model, with the preprocessing techniques of normalization, augmentation, and class balancing to make the AI model more robust in different skin tones and under various lighting conditions. EfficientNet-Lite and Optimized MobileNetV3 architectures recorded a competitive accuracy of 8790 percent with a much smaller model size to deploy efficiently on mobile devices. To simulate the actual reality of usage, the development and testing was done in Visual Studio Code upon physical Android devices. Feedbacks on the system were reported to be very satisfying by pilots in terms of the clarity, responsiveness, and user-friendliness. In general, the app confirms the efficiency of integrating edge detection with chatbots into a fast, convenient and privacy-aware solution that can raise skin-health awareness in the early stages and increase the availability of preliminary dermatological consultation, particularly, in communities with lower resources and limited access to them. Ms. Rabia Masood 2026-01-01T00:00:00Z