Laboratories may produce thousands of images during the experimentation phase, capturing and storing them across various devices. Keeping a comprehensive record of these files so that they can be retrieved when needed can be difficult and requires effective storage and file management. If researchers use the wrong image, it can lead to confusion and potential retraction. Here Dror Kolodkin-Gal, PhD, founder of automated image integrity tool Proofig AI, explains how image issues occur, the consequences of including image issues and how lab managers, principal investigators (PIs), and their support staff can protect the integrity of their work.
The importance of image integrity
Scientific images are commonly shared in lab research and manuscripts to convey complex data, observations, and results. To ensure all the data shared in these figures positively impacts scientific progress, and by extension benefits the wider community, it is crucial for researchers, lab managers, PIs, and editors to uphold data integrity.
Get training in Lab Quality and earn CEUs.
One of over 25 IACET-accredited courses in the Academy.
Lab Quality course
Accurate images help prevent misinformation, misinterpretation, and potential reputational damage, safeguarding both scientific rigor and the institution associated with the lab. However, despite its importance, maintaining image integrity poses significant challenges and remains a common issue in lab research.
The problem with image integrity
Over the last few years, there have been high-profile cases of potential scientific integrity issues that highlight the severity of the current replication crisis. Presidents of reputable academic institutions have resigned after investigations into their research and landmark studies that influenced public treatment are now being investigated1-2, questioning the validity of years of studies based on the original results. Reports of potential misconduct such as these can negatively affect the reputation of the institutions and scientists associated with the lab, which may later impact public trust and decrease funding opportunities to the institute.
PIs should do everything they can to prevent the falsification of research or instances of image manipulation and errors. While preventing misconduct is a key issue, PIs should also be aware that most image integrity issues are found to be unintentional. This was evidenced during a trial that ran from January 2021 to May 2022, where to the AACR screened 1,367 papers accepted for publication3. Of those, 208 papers required author contact to clear up issues such as mistaken duplications, and only four papers were withdrawn. In almost all cases (204 out of 208), there was no evidence of intentional image manipulation.
Protecting the laboratory
By integrating systematic quality assurance protocols, PIs and lab managers can ensure that images are meticulously managed and checked for integrity before publication. PIs can identify any signs of misconduct, as well as prevent any unintentional mistakes from leaving the lab by identifying and correcting issues ahead of submission. This helps avoid costly retractions and maintain the integrity of trustworthiness of the research they share, mentoring others in the lab on how to avoid future image integrity issues.
First, PIs should encourage best practice when capturing images, establishing clear protocols for labeling, storing, and managing images. Training every researcher on proper image management and image integrity, and mentoring them as they gather evidence for research will help reduce any instances of image integrity issues. This includes regular training sessions on proper image management and image integrity, ensuring that every researcher understands the significance of these practices and is equipped to adhere to them. Monthly meetings can be valuable for discussing and sharing new tools and strategies for maintaining lab standards, which not only enhances organization but also raises awareness about the importance of image integrity.
Additionally, PIs can leverage tools like to streamline data management and ensure that images are securely stored and easily accessible. For example, electronic lab notebooks (ELNs) provide a structured environment for maintaining research records, minimizing the risk of data loss or unauthorized alterations. Bringing in external experts annually to discuss the importance of research integrity and the available tools can reinforce the lab's commitment to high standards. Integrating AI-powered solutions for automated text analysis for integrity issues and image proofing further ensures that research outputs meet ethical guidelines, reducing the likelihood of errors or manipulations before publication.
PIs and lab managers can also make image integrity checks a key part of the research workflow. Checking images before grant and publication submissions and following acceptance of manuscripts can prevent issues during or after the review process. Scientific images can be incredibly detailed and complex, making it challenging to review accurately by eye. This issue, coupled with the fact that PIs must juggle various other responsibilities, makes meticulously checking each image time-consuming and demanding. Often, it is impossible for PIs to detect issues in images with the naked eye due to the need to perform hundreds of thousands of different checks on images in a paper, necessitating the assistance of specialized software. Software tools can help to automate the image integrity process, scanning manuscripts in minutes to flag any potential duplications or manipulations. PIs can then review the report and investigate any flagged issues further, reducing the likelihood of errors in a much shorter time frame compared with manual checks.
For PIs and lab managers, maintaining image integrity is not just about protecting the reputation of individual researchers or the lab, but also about contributing to the broader integrity of the scientific community. By sharing image integrity best practice and introducing tools that can help more accurately detect issues ahead of submission, PIs leading the lab can play an integral role in safeguarding the scientific record, ensuring any research shared is accurate and credible.
References:
1. https://www.nytimes.com/2023/07/19/us/stanford-president-resigns-tessier-lavigne.html
