Evaluating Large Language Models for Radiology Natural Language Processing

The rise of large language models (LLMs) has marked a pivotal shift in the field of natural language processing (NLP). LLMs have revolutionized a multitude of domains, and they have made a significant impact in the medical field. Large language models are now more abundant than ever, and many of these models exhibit bilingual capabilities, proficient in both English and Chinese. However, a comprehensive evaluation of these models remains to be conducted. This lack of assessment is especially apparent within the context of radiology NLP. This study seeks to bridge this gap by critically evaluating thirty two LLMs in interpreting radiology reports, a crucial component of radiology NLP. Specifically, the ability to derive impressions from radiologic findings is assessed. The outcomes of this evaluation provide key insights into the performance, strengths, and weaknesses of these LLMs, informing their practical applications within the medical domain.

Exploring the Trade-Offs: Unified Large Language Models vs Local Fine-Tuned Models for Highly-Specific Radiology NLI Task

Recently, ChatGPT and GPT-4 have emerged and gained immense global attention due to their unparalleled performance in language processing. Despite demonstrating impressive capability in various open-domain tasks, their adequacy in highly specific fields like radiology remains untested. Radiology presents unique linguistic phenomena distinct from open-domain data due to its specificity and complexity. Assessing the performance of large language models (LLMs) in such specific domains is crucial not only for a thorough evaluation of their overall performance but also for providing valuable insights into future model design directions: whether model design should be generic or domain-specific. To this end, in this study, we evaluate the performance of ChatGPT/GPT-4 on a radiology NLI task and compare it to other models fine-tuned specifically on task-related data samples. We also conduct a comprehensive investigation on ChatGPT/GPT-4's reasoning ability by introducing varying levels of inference difficulty. Our results show that 1) GPT-4 outperforms ChatGPT in the radiology NLI task; 2) other specifically fine-tuned models require significant amounts of data samples to achieve comparable performance to ChatGPT/GPT-4. These findings demonstrate that constructing a generic model that is capable of solving various tasks across different domains is feasible.

DeID-GPT: Zero-shot Medical Text De-Identification by GPT-4

The digitization of healthcare has facilitated the sharing and re-using of medical data but has also raised concerns about confidentiality and privacy. HIPAA (Health Insurance Portability and Accountability Act) mandates removing re-identifying information before the dissemination of medical records. Thus, effective and efficient solutions for de-identifying medical data, especially those in free-text forms, are highly needed. While various computer-assisted de-identification methods, including both rule-based and learning-based, have been developed and used in prior practice, such solutions still lack generalizability or need to be fine-tuned according to different scenarios, significantly imposing restrictions in wider use. The advancement of large language models (LLM), such as ChatGPT and GPT-4, have shown great potential in processing text data in the medical domain with zero-shot in-context learning, especially in the task of privacy protection, as these models can identify confidential information by their powerful named entity recognition (NER) capability. In this work, we developed a novel GPT4-enabled de-identification framework ("DeID-GPT") to automatically identify and remove the identifying information. Compared to existing commonly used medical text data de-identification methods, our developed DeID-GPT showed the highest accuracy and remarkable reliability in masking private information from the unstructured medical text while preserving the original structure and meaning of the text. This study is one of the earliest to utilize ChatGPT and GPT-4 for medical text data processing and de-identification, which provides insights for further research and solution development on the use of LLMs such as ChatGPT/GPT-4 in healthcare. Codes and benchmarking data information are available at https://github.com/yhydhx/ChatGPT-API.

MUI-TARE: Multi-Agent Cooperative Exploration with Unknown Initial Position

Multi-agent exploration of a bounded 3D environment with unknown initial positions of agents is a challenging problem. It requires quickly exploring the environments as well as robustly merging the sub-maps built by the agents. We take the view that the existing approaches are either aggressive or conservative: Aggressive strategies merge two sub-maps built by different agents together when overlap is detected, which can lead to incorrect merging due to the false-positive detection of the overlap and is thus not robust. Conservative strategies direct one agent to revisit an excessive amount of the historical trajectory of another agent for verification before merging, which can lower the exploration efficiency due to the repeated exploration of the same space. To intelligently balance the robustness of sub-map merging and exploration efficiency, we develop a new approach for lidar-based multi-agent exploration, which can direct one agent to repeat another agent's trajectory in an \emph{adaptive} manner based on the quality indicator of the sub-map merging process. Additionally, our approach extends the recent single-agent hierarchical exploration strategy to multiple agents in a \emph{cooperative} manner by planning for agents with merged sub-maps together to further improve exploration efficiency. Our experiments show that our approach is up to 50\% more efficient than the baselines on average while merging sub-maps robustly.

Emulating Quantum Dynamics with Neural Networks via Knowledge Distillation

High-fidelity quantum dynamics emulators can be used to predict the time evolution of complex physical systems. Here, we introduce an efficient training framework for constructing machine learning-based emulators. Our approach is based on the idea of knowledge distillation and uses elements of curriculum learning. It works by constructing a set of simple, but rich-in-physics training examples (a curriculum). These examples are used by the emulator to learn the general rules describing the time evolution of a quantum system (knowledge distillation). The goal is not only to obtain high-quality predictions, but also to examine the process of how the emulator learns the physics of the underlying problem. This allows us to discover new facts about the physical system, detect symmetries, and measure relative importance of the contributing physical processes. We illustrate this approach by training an artificial neural network to predict the time evolution of quantum wave packages propagating through a potential landscape. We focus on the question of how the emulator learns the rules of quantum dynamics from the curriculum of simple training examples and to which extent it can generalize the acquired knowledge to solve more challenging cases.

Autonomous Exploration Development Environment and the Planning Algorithms

Autonomous Exploration Development Environment is an open-source repository released to facilitate the development of high-level planning algorithms and integration of complete autonomous navigation systems. The repository contains representative simulation environment models, fundamental navigation modules, e.g., local planner, terrain traversability analysis, waypoint following, and visualization tools. Together with two of our high-level planner releases -- TARE planner for exploration and FAR planner for route planning, we detail usage of the three open-source repositories and share experiences in the integration of autonomous navigation systems. We use DARPA Subterranean Challenge as a use case where the repositories together form the main navigation system of the CMU-OSU Team. In the end, we discuss a few potential use cases in extended applications.

FAR Planner: Fast, Attemptable Route Planner using Dynamic Visibility Update

We present our work on a fast route planner based on visibility graph. The method extracts edge points around obstacles in the environment to form polygons, with which, the method dynamically updates a global visibility graph, expanding the visibility graph along with the navigation and removing edges that become occluded by dynamic obstacles. When guiding a vehicle to the goal, the method can deal with both known and unknown environments. In the latter case, the method is attemptable in discovering a way to the goal by picking up the environment layout on the fly. We evaluate the method using both ground and aerial vehicles, in simulated and real-world settings. In highly convoluted unknown or partially known environments, our method is able to reduce travel time by 13-27% compared to RRT*, RRT-Connect, A*, and D* Lite, and finds a path within 3ms in all of our experiments.

Dynamic Channel: A Planning Framework for Crowd Navigation

Real-time navigation in dense human environments is a challenging problem in robotics. Most existing path planners fail to account for the dynamics of pedestrians because introducing time as an additional dimension in search space is computationally prohibitive. Alternatively, most local motion planners only address imminent collision avoidance and fail to offer long-term optimality. In this work, we present an approach, called Dynamic Channels, to solve this global to local quandary. Our method combines the high-level topological path planning with low-level motion planning into a complete pipeline. By formulating the path planning problem as graph searching in the triangulation space, our planner is able to explicitly reason about the obstacle dynamics and capture the environmental change efficiently. We evaluate efficiency and performance of our approach on public pedestrian datasets and compare it to a state-of-the-art planning algorithm for dynamic obstacle avoidance.

DoraPicker: An Autonomous Picking System for General Objects

Robots that autonomously manipulate objects within warehouses have the potential to shorten the package delivery time and improve the efficiency of the e-commerce industry. In this paper, we present a robotic system that is capable of both picking and placing general objects in warehouse scenarios. Given a target object, the robot autonomously detects it from a shelf or a table and estimates its full 6D pose. With this pose information, the robot picks the object using its gripper, and then places it into a container or at a specified location. We describe our pick-and-place system in detail while highlighting our design principles for the warehouse settings, including the perception method that leverages knowledge about its workspace, three grippers designed to handle a large variety of different objects in terms of shape, weight and material, and grasp planning in cluttered scenarios. We also present extensive experiments to evaluate the performance of our picking system and demonstrate that the robot is competent to accomplish various tasks in warehouse settings, such as picking a target item from a tight space, grasping different objects from the shelf, and performing pick-and-place tasks on the table.

Analyzing the Utility of a Support Pin in Sequential Robotic Manipulation

Pick-and-place regrasp is an important manipulation skill for a robot. It helps a robot accomplish tasks that cannot be achieved within a single grasp, due to constraints such as kinematics or collisions between the robot and the environment. Previous work on pick-and-place regrasp only leveraged flat surfaces for intermediate placements, and thus is limited in the capability to reorient an object. In this paper, we extend the reorientation capability of a pick-and-place regrasp by adding a vertical pin on the working surface and using it as the intermediate location for regrasping. In particular, our method automatically computes the stable placements of an object leaning against a vertical pin, finds several force-closure grasps, generates a graph of regrasp actions, and searches for the regrasp sequence. To compare the regrasping performance with and without using pins, we evaluate the success rate and the length of regrasp sequences while performing tasks on various models. Experiments on reorientation and assembly tasks validate the benefit of using support pins for regrasping.