Overview About Achievion Solutions

Client A prominent medical equipment vendor specializing in innovative diagnostic and monitoring devices for healthcare providers. Objective Develop a highly accurate, user-friendly cardiac monitoring system suitable for both clinical environments and home use. The key challenge was creating a solution capable of interpreting complex electrocardiogram (ECG) signals to detect a wide range of cardiac anomalies, ensuring reliable, real-time analysis. Solution We delivered an advanced AI-powered software solution that transforms raw ECG data into actionable insights, enhancing diagnostic precision and patient care. This intelligent cardiac monitoring platform incorporates the latest machine learning techniques and real-time adaptability. Key Features: Deep Neural Networks for ECG Analysis Utilized Convolutional Neural Networks (CNNs) trained on extensive annotated ECG datasets. Enabled precise detection and classification of a broad spectrum of cardiac conditions. Real-Time Data Processing Designed for instant ECG analysis, providing healthcare professionals with timely alerts for potential cardiac events. Improved patient outcomes through proactive intervention. Patient-Specific Model Adaptation Integrated adaptive AI models that evolve with each patient’s unique cardiac patterns. Enhanced accuracy and reliability through personalized, evolving monitoring. Result The AI-powered heart monitoring system significantly advanced cardiac care by offering: Continuous and precise heart health monitoring. Actionable insights for healthcare providers and patients. Personalized adaptability, improving effectiveness over time. This innovative solution stands as a robust tool for personalized healthcare, ensuring better diagnostic outcomes and improved patient experiences.

Client US-based hospital looking to leverage AI to optimize its patient care process. Objective Client tasked our team with developing an AI-driven system that can identify the most promising cancer patients for treatment in a hospital overwhelmed by a patient load ten times greater than its capacity. The AI will also predict the probability of successful treatment for each patient. Solution Our approach to this extensive project was divided into three key phases: Phase 1: Data extraction and preparation The hospital stores patient information in the Cerner EMR system, but much of this data exists in unstructured PDF format. This format hinders the effective use of AI. Therefore, the initial crucial step is to extract patient data from these PDFs and make it accessible for AI applications. To tackle this challenge, we implemented the following steps: Exported PDF data: we transferred patient data from Cerner to AWS Data Lake for centralized storage. Text conversion: utilized AWS Textract to convert PDF content into plain text format. Data cleansing: cleaned the raw data using the Talend ETL tool and stored it in AWS Redshift. Clinical concept extraction: applied AWS Comprehend and AWS Comprehend Medical to identify and extract clinical concepts from the data. For cases where AWS Comprehend Medical provided insufficient results, we developed and managed a custom AI model using BioBERT, a neural network trained on medical texts. This model was trained on the hospital’s historical data to enhance accuracy. Phase 2: AI-powered patient admissions module development Having enriched patient data, we moved forward with developing a smart patient admissions module: Data preprocessing: used AWS Glue for data preprocessing, cleansing, and outlier detection. Applied techniques to handle missing data. Feature engineering: identified and refined relevant features from the raw data for model training. Model training: trained a LightGBM model, a powerful algorithm for predicting patient outcomes and treatment success. Evaluated the model’s performance through cross-validation and fine-tuned its parameters. Visualization: created slick user interface to illustrate how different features influenced model predictions, making the results easier to understand. Deployment: deployed and managed the model using AWS SageMaker, ensuring it was rigorously tested and regularly updated with new data to maintain its accuracy and relevance. Phase 3: Treatment success prediction module development In the final phase, we focused on developing a module for predicting treatment success and determining the optimal timing for hospice transitions: Data processing: used AWS Glue for comprehensive data preprocessing, including the addition of temporal features like treatment duration and progression markers. Model development: trained a LightGBM model combined with survival analysis techniques such as the Kaplan-Meier estimator and Cox proportional hazards model to predict treatment success. Performance evaluation: evaluated the model’s performance through cross-validation and adjusted its parameters for optimal results. Visualization: utilized tools like SHAP and Seaborn to visualize feature importance and model predictions, providing clear insights for decision-making. Deployment: deployed the model using AWS SageMaker, ensuring rigorous testing and regular updates with new data to keep the model accurate and relevant. Result Our innovative solution tackled the hospital’s complex challenges and delivered measurable improvements: Developed and implemented a cutting-edge smart patient admissions module, enhancing the hospital’s capacity to manage high patient volumes efficiently. Enabled the hospital to handle an increased number of patients without compromising the quality of care, optimizing both patient flow and resource allocation. Utilized advanced AI models to predict treatment outcomes with greater accuracy, supporting better decision-making for patient care. Optimized hospice care decisions through AI-driven insights, leading to more personalized and effective patient care strategies. These strategic advancements not only addressed the immediate challenges but also set the stage for future improvements in hospital operations and patient care.

AI-powered dataset management solution reduced operational costs by 29%. Client: Healthcare consulting company specializing in data analytics and support of the medical research community, set a goal of automating its dataset management process with AI technology. Objective: The client was looking to leverage AI to automate dataset lookup and import, metadata harvesting, natural language queries execution, and other tasks. On top of that they decided to build and productize the web-based platform to offer this solution to other data analytics companies. Solution: Achievion team started the project by analyzing all existing dataset variations and collecting platform requirements. Once all details were clarified, we designed scalable microservice-based solution architecture. Then our engineering team developed and delivered the web-based platform using ACHIEVE methodology. The product includes the following features: Search database using filters, keywords, and metadata variables View and analyze search results Compare datasets and variables Admin interface for dataset management We established the data collection mechanism to enable AI/ML models training and AI-powered capabilities implementation: Intelligent data search Automated metadata harvesting Metadata variable summarization Natural language search queries Predictive analytics to determine potential gaps in the database and inform future dataset selection needs Hyper-personalization of the search results according to individual user preferences Result: The AI-powered dataset analytics platform delivered anticipated improvements in efficiency and cost-effectiveness to the client. Reduced operational costs: through the automation of tasks such as dataset management, including lookup, import, metadata harvesting, and execution of natural language queries, the client witnessed a considerable decrease in operational costs, amounting to approximately 29%. Increased productivity: this streamlined approach to dataset management translated into a 38% surge in productivity among data analytics teams. Consequently, they could allocate more time to strategic initiatives and support for medical research. New revenue stream: opting for the development of a web-based platform proved to be a lucrative decision, as it paved the way for a new revenue stream for the client. The anticipated annual revenue growth is estimated at 31%.

Our development team has engineered an innovative Interior Design AI System that harnesses the capabilities of natural language processing and computer vision. This system enables users to seamlessly generate, customize, and envision interior designs with ease. The system comprises the following modules: Generation of new interior designs using user prompts and existing layouts Real-time editing of designs with simplicity Material recommendations and selection based on a comprehensive catalog Integration with third-party cost catalogs Accurate calculation of price quotes

Pay4SchoolStuff.com is a payment and information management solution for leaders and members of school community organizations. Achivion became the primary IT vendor for this product in 2014. We improved usability, stabilized infrastructure performance and continue to enhance platform features to help them remain #1 in their niche market.

ItsOnMe is a SaaS-based e-Gifting platform for small businesses. It is designed to efficiently integrate e-Gifting into the business model. Users can send company e-gift card customized with a personal picture or video, and delivered via email, Facebook, or phone. Achievion's mobile team built this platform from scrath and delivered revolutionary electronic gifting ideas leveraging latest technology trends.

The National Veterans Legal Services Program (NVLSP) is a non-profit organization working on behalf of veterans and active duty military personnel to ensure that they receive the benefits they are entitled to having become disabled while performing military service for our country.   NVLSP was looking to develop web and mobile apps that would help veterans and active duty personnel determine what disability benefits would be available to them.  The solution allows administrators create questionnaire landing pages, add questions and more. The web app version of the Veterans Benefits Questionnaire app makes the questionnaire accessible to users, allowing them to answer questions and download their results as a PDF file or have the questionnaire results sent to them via email. The mobile app offers similar features as the web app, while optimizing the questionnaire experience for mobile. With both a web and mobile app, we've helped NVLSP to make it easier for their potential clients to find out about their disability benefits. For NVLSP administrators, the new apps also streamline the process of collecting important questionnaire data and reviewing submitted questionnaires.

Telnet, Inc. is a consulting firm providing complete Telecom and IT solutions globally. They were looking for a solution that would drastically reduce the EME testing project duration from several weeks to several days. Achievion build a platform that successfully achieved the goal and, moreover, delivered a SaaS application which they can sell on subscription basis to other telecom companies.

Blockchain Information Exchange (BIX) Systems company developed a next generation secure distributed ledger based on blockchain architecture. Their infrastructure overcomes the traditional blockchain issues of trust, security, scalability and offers a platform for building enterprise-grade distributed applications. BIX team is comprised of world-class academic, computer security and blockchain experts who developed the backbone of the platform. In 2017 their solution started to get traction among financial, real estate, healthcare startups and organizations looking to leverage the benefits of the blockchain technology. To meet the growing demand, BIX decided to partner with a software development agency experienced in building sophisticated blockchain-based custom applications. After a series of introductory meetings Achievion was selected as a partner team. We analyzed client’s microservices-based platform architecture, requirements for prospective applications and built a common Template module that expedites future apps development by an order of magnitude. Once the Template module was developed, we designed and delivered a number of custom applications built on top of that module in a very short period of a few months.

G&J Moving is a Virginia based moving and storage company making over 2000 local and long distance moves per year. They provide end-to-end moving services to both residential and commercial clients. Client was looking to design and develop a web-based software to optimize long-distance shipment management, automate delivery planning and increase routes efficiency. We created a cloud-based web application that integrates with multiple freight forwarding services, calculates most profitable delivery routes and delivers them via user-friendly interface to the client’s operations manager for execution. Features include: Automated shipments information ingest from various freight forwarding services Ability to recognize over 100,000 variables Delivery schedule planning using AI-powered proprietary algorithm Integration with client’s in-house transport management system Easy to use web user interface for accessing all components of the software Client’s operations department got a 30-40% boost in long-distance delivery planning efficiency. Route dispatcher now spends way less time looking for the best routes and instead focuses on quality shipment delivery execution.

A proprietary security solutions company with a focus on the events industry was seeking to enhance security and surveillance at events using powerful facial-recognition technology. After much forethought and brainstorming as well as several trial runs, the team at Achievion developed a biometric application for face recognition solution that makes it possible to extract features from a person’s photos by means of computer vision.

Payment platform specializing in handling transactions for companies in travel industry had a need to identify fraudulent transactions and reduce fraud related losses. Achievion data science experts analyzed the client’s data lake that aggregated financial data coming from multiple sources. The team has also reviewed the existing transaction validation engine to better understand the nature of the fraudulent activities. Once the analysis phase was completed, we developed and trained the Machine Learning model to detect suspicious transactions within several seconds after posting. The continuous monitoring and improvement process was put in place to keep the model up to date going forward. Our engineering team built the API on top of the model and integrated the software with client’s payment processing engine. Our software significantly increased the number of detected fraudulent transactions and achieved 95% accuracy. The client’s manual transaction review process got simplified which brought substantial cost savings to the company.

Forte is a video streaming platform that offers its users live and on-demand boutique fitness classes from top studios worldwide. They lost their CTO and entire engineering team and desperately needed new technology partner to rescue the platform and transform it from a classic marketplace website to a SaaS ecosystem. Achievion team was able to quickly pick up where the former tech team left off. They were able to get platform up and running within several days. The team built a scalable SaaS infrastructure and enhanced the platform with the following features: Class leaderboard to allow users compete with each other in real-time  2-way video capability to enable class participants communicate in a Zoom-like way  Connectivity with top wearable devices (Apple Watch, FitBit, Garmin, Samsung Watch, MyZone) for instant workout performance feedback White-label module to deliver custom web and mobile Forte applications to the clients  Integration with multiple club management software vendors such as MindBody, MarianaTek, ABC Fitness and Daxco to ensure smooth integration between the white-label SaaS system and existing studio websites iOS and Android apps The team became a trusted partner to Forte and built a scalable SaaS infrastructure serving top fitness chains across US and abroad.

Drug effectiveness prediction software Pharmaceutical company specializing in oncology, gastroenterology, and neuroscience drugs development was looking to develop AI-powered solution to determine the probability of negative response to components of several drugs developed to treat lung cancer, epilepsy and measles. Our team collected data samples from multiple sources including company R&D department, observational studies, EHRs, hospitals and public databases. We then cleaned and labeled the data for effective analysis. Using various Machine Learning algorithms, techniques and frameworks, our data scientists built a model that was able to predict drug effectiveness across a range of parameters. On top of that, the solution can suggest alternative clinical trial approaches in case of poor treatment outcomes. Our AI-powered software increased the treatment efficiency of several widely spread diseases by decreasing the side effects risk.

Computer vision powered retail traffic analytics Technology company providing data-driven solutions to the brick-and-mortar retail sector was looking to build a real-time mall traffic monitoring solution by analyzing camera footage to help operation teams manage store occupancy efficiently and capitalize on consumer insights. Our team performed a broad analysis of the video footage collected from several shopping malls. Then we designed intuitive platform user interface and delivered the software that includes the following features: Visitors count and pathways tracking Heatmaps generation across various metrics such as number of people, dwell time, traffic journeys and others Store lines monitoring and management Social distancing and store occupancy monitoring MAG (mood/age/gender) visitor profiling Marketing campaign performance measurements Our software was deployed to several shopping malls and produced a multitude of insightful data points about visitors. This information allowed individual store owners to deliver personalized offerings to their clients and achieve higher retention rates.

A California-based startup in the VC investment industry spent over a year analyzing profiles of a wide range of startups to see if there is a way to predict the success of a given company before pouring money into it. It turned out that data analytics alone was not able to produce accurate results. So they set their eyes on the emerging field of Artificial Intelligence to get more help. After investing a significant amount of time and energy, they came up with a proprietary AI-based approach that yielded much better outcomes. Now they needed a company who could turn their mathematical algorithms into a product. Using recurrent neural network framework as the brain of the solution, we turned the client’s mathematical algorithms into an elegant AI based product. The proprietary AI-based approach to predicting the success of investments yielded much better outcomes than data analytics alone. The AI based product makes it easy for the end-user to make business decisions for various investment scenarios.

Project and document management systems (DMS) provider offering software products for construction and real estate industries was looking to streamline manual document entry process for the end users of the client’s DMS. Our data science experts reviewed samples of the typical documents that get entered into the system and built the Machine Learning model that broke them down into categories based on the OCRed content. Then the software engineering team wrapped the model into a group of APIs deployed to the AWS cloud and integrated the solution with client’s document management system.

Boston-based home services marketplace and social network connecting customers and pros through project proposal, bid, escrow payment, and mutual rating. Client's team evaluated the home services technology environment and identified a unique niche on the intersection of Angie’s list, Upwork and Instagram. After consulting with C-level executives from Airbnb and other successful “startups” they decided to launch the next-gen platform and change the home improvement world. After rigorous selection process Achievion was picked to develop the platform that includes responsive web application as well as iOS and Android mobile apps. Our team developed a technical specification, designed modern UI and delivered the first version of the platform in 6 months. Features include: Home projects dashboard Sophisticated bidding system Multi-step project delivery workflow Contractor verification in Checkr Homeowner service payments and distribution to assigned contractor via Stripe Mutual rating module Basic social network called “My Neighborhood” Platform Admin dashboard During the first month of the platform roll out in the Boston market, the team has registered hundreds of customers and service professionals and jobs started to pour in. By the end of the first year after launch, the platform attracted over 50,000 users and helped customers complete over 2000 projects.

Just Answer is an online answer engine that connects visitors to experts in various disciplines. It uses the paid-advice model. The client was seeking an AI solution to route questions to human experts and improve the speed and efficiency of the ‘question-answer’ process. Using AI, machine learning and natural language processing, the Just Answer Chatbot was developed to serve as an intelligent routing engine. As a result, questions are routed faster, communication between user and expert is more efficient. The improved efficiency and speed have resulted in increased conversions which allows the online expert question and answer website to stay relevant and grow.